A multi-unmanned platform-oriented situation information fusion method and device
By using trajectory initialization, trajectory extrapolation, and filtering, combined with autocorrelation calculation and singular value decomposition, the noise and discontinuity problems of unmanned platform trajectory data are solved, enabling accurate fusion and real-time processing of multi-source trajectory data and supporting collaborative operations of unmanned platforms.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INST OF LOGISTICS SCI & TECH ACAD OF SYST ENG ACAD OF MILITARY SCI
- Filing Date
- 2025-06-30
- Publication Date
- 2026-08-04
AI Technical Summary
Existing technologies face challenges in processing trajectory data from unmanned platforms, including noise and discontinuity, multiple sources, complexity, real-time nature, and dynamism. These challenges make trajectory data processing difficult and affect analysis results and real-time decision-making.
By employing trajectory initialization, trajectory extrapolation, and trajectory filtering, combined with autocorrelation calculation, singular value decomposition, and Gaussian kernel function smoothing, data preprocessing and trajectory estimation are performed to achieve the fusion of multi-source sensor information and spatial feature registration.
It effectively removes noise and outliers from trajectory data, fills in missing data, improves the accuracy and continuity of trajectory data, achieves effective fusion and spatial consistency of multi-source trajectory data, and supports real-time decision-making and control.
Smart Images
Figure CN122506537A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of unmanned swarm technology and information processing, specifically to a method and apparatus for situational information fusion for multiple unmanned platforms. Background Technology
[0002] In the field of trajectory monitoring and analysis for modern unmanned platforms (such as drones and autonomous vehicles), accurate acquisition and processing of target trajectory information is crucial. However, existing technologies face numerous technical challenges when processing trajectory data from unmanned platforms:
[0003] (I) Noise and Discontinuity Issues in Trajectory Data: During operation, unmanned platforms often experience noise and discontinuities in their target position coordinates due to limitations in sensor accuracy, environmental interference (such as electromagnetic interference and terrain occlusion), and insufficient performance of data acquisition equipment. For example, when a UAV flies in a complex urban environment, building obstructions may cause temporary loss or deviation of its position data, resulting in jumps or abnormal fluctuations in the trajectory data. This not only affects the visualization of the trajectory but may also lead to serious errors in subsequent analyses (such as path planning and behavior prediction).
[0004] (II) The Multi-Source and Complexity of Trajectory Data: In collaborative operation scenarios involving multiple unmanned platforms, the trajectory data of each platform is typically collected by different sensors, which may have different sampling rates, accuracies, and data formats. For example, some drones may be equipped with high-precision GPS modules, while others may rely solely on low-precision inertial measurement units (IMUs). This diversity of multi-source data makes it difficult to directly fuse and analyze trajectory data. Furthermore, the motion patterns of unmanned platforms are complex and diverse, potentially including linear motion, curvilinear motion, acceleration, and deceleration, further increasing the difficulty of trajectory data processing.
[0005] (iv) Real-time and dynamic issues of trajectory data: The motion trajectory of unmanned platforms is dynamically changing, requiring real-time processing and analysis of trajectory data to support real-time decision-making and control. However, existing technologies often suffer from slow processing speed and high latency when processing trajectory data. For example, while some trajectory filtering algorithms can effectively remove noise, their computational complexity is high, making it difficult to meet real-time requirements. Furthermore, the motion state of unmanned platforms may change suddenly (such as emergency obstacle avoidance or sudden acceleration), and existing technologies struggle to adapt quickly to such dynamic changes, resulting in delayed trajectory data processing results. Summary of the Invention
[0006] This invention mainly addresses the problems of discontinuity, multi-source nature and complexity, spatial consistency, real-time nature and dynamism of trajectory data in existing unmanned platform situation processing technologies. This invention discloses a situation information fusion method and device for multiple unmanned platforms.
[0007] In a first aspect, this invention discloses a situational information fusion method for multiple unmanned platforms, comprising:
[0008] S1, acquire geospatial data information and a set of multi-source sensor detection information from the unmanned platform; the set of multi-source sensor detection information includes multi-sensor detection information from each unmanned platform; the multi-sensor detection information includes infrared detection information and radar detection information;
[0009] S2, perform trajectory fusion estimation processing on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform; the trajectory information set includes the trajectory information of each unmanned platform;
[0010] S3, load the set of flight track information into the geospatial data information to obtain the situation information of the unmanned platform; and visualize the situation information.
[0011] The process of performing trajectory fusion estimation on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform includes:
[0012] S21, perform data preprocessing on the multi-source sensor detection information set to obtain a preprocessed information set;
[0013] S22, Perform trajectory estimation processing on the preprocessed information set to obtain the trajectory information set of the unmanned platform.
[0014] The step of preprocessing the multi-source sensor detection information set to obtain a preprocessed information set includes:
[0015] S211, perform data cleaning processing on the multi-source sensor detection information set to obtain the first dataset;
[0016] S212, Smooth the first dataset to obtain the second dataset;
[0017] S213, perform category discrimination processing on the second dataset to obtain a preprocessed information set.
[0018] The process of performing trajectory estimation on the preprocessed information set to obtain the trajectory information set of the unmanned platform includes:
[0019] S221, The preprocessed information set is processed once to obtain the target motion parameter information set;
[0020] S222, perform secondary processing on the target motion parameter information set to obtain the target trajectory information set;
[0021] S223, the target trajectory information set is processed three times to obtain the flight track information set.
[0022] The secondary processing of the target motion parameter information set to obtain the target trajectory information set includes:
[0023] S2221, Perform trajectory initialization, trajectory extrapolation and trajectory filtering on the target position coordinate information of each unmanned platform in the target motion parameter information set to obtain the trajectory information set of the unmanned platform;
[0024] S2222: Spatial feature registration is performed on the track information set of each unmanned platform to obtain the corresponding registered track information set.
[0025] The process of spatial feature registration of the trajectory information set for each unmanned platform to obtain the corresponding registered trajectory information set includes:
[0026] Clustering is performed on the track information set of each unmanned platform to obtain category information and track information included in each category;
[0027] For each category of track information, a first fusion calculation is performed to obtain the fused track information for each category;
[0028] By utilizing the fused track information from all unmanned platforms, a registration track information set is constructed; the registration track information set includes fused track information of all categories for each unmanned platform.
[0029] The first fusion calculation process is performed on the track information of each category to obtain the fused track information of each category, including:
[0030] For each category of track information, a track matrix is represented; the three row vectors of the track matrix are the sequence of track position coordinates at all times, representing the x-axis, y-axis, and z-axis positions.
[0031] Autocorrelation is calculated on the row vectors of the track matrix to obtain the autocorrelation matrix; each row vector of the autocorrelation matrix is the autocorrelation sequence of the row vectors corresponding to the track matrix.
[0032] Singular value decomposition is performed on the autocorrelation matrix to obtain the singular value matrix;
[0033] Extract the diagonal elements of the singular value matrix to obtain the singular vector;
[0034] Linear fitting is performed on the elements and element index values of the singular vector to obtain the singularity discrimination polynomial;
[0035] Using the singularity discriminant polynomial, the index values of all elements of the singular vector are calculated and processed to obtain the fitted singular vector;
[0036] Based on the autocorrelation matrix, the trajectory matrix is transformed to obtain the transformation matrix;
[0037] The track matrix is smoothed using the transformation matrix and fitted singular vectors of all track information of a category to obtain the fused track matrix; the column vectors of the fused track matrix are the three-dimensional position coordinates of the fused track information at a certain moment.
[0038] The fused track matrix is determined to be the fused track information of the category.
[0039] A second aspect of this invention discloses a situational information fusion device for multiple unmanned platforms, the device comprising:
[0040] Memory containing executable program code;
[0041] A processor coupled to the memory;
[0042] The processor calls the executable program code stored in the memory to execute the situation information fusion method for multiple unmanned platforms.
[0043] In a third aspect of this invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, and when the computer instructions are invoked by a computer, they are used to execute the situation information fusion method for multiple unmanned platforms.
[0044] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the situational information fusion method for multiple unmanned platforms.
[0045] The beneficial effects of this invention are as follows:
[0046] This invention, through track initialization, track extrapolation, and track filtering, effectively removes noise and outliers from trajectory data, filling in missing data and thus improving the accuracy and continuity of the trajectory data. For example, for the problem of transient position data loss when UAVs fly in complex environments, track extrapolation can predict the approximate location of the lost portion based on existing trajectory information, while track filtering can further smooth the trajectory and eliminate fluctuations caused by noise. This makes the trajectory data more reliable, providing a solid foundation for subsequent analysis and applications.
[0047] This invention performs clustering and fusion calculations on the trajectory information sets of each unmanned platform, effectively integrating trajectory data from different sources and with varying precision. By representing the trajectory information of each category as a trajectory matrix and performing a series of processes such as autocorrelation calculation and singular value decomposition, the fused trajectory information is finally obtained. This process not only considers the spatial characteristics of the trajectory data but also further optimizes the fusion result through methods such as singular discriminant multinomials and fitted singular vectors. For example, in multi-UAV collaborative missions, even if there are significant differences in the trajectory data of different UAVs, this invention can fuse them into a unified registered trajectory information set, achieving effective fusion of multi-source trajectory data.
[0048] This invention performs spatial feature registration on the trajectory information sets of each unmanned platform, enabling spatial alignment and matching of trajectory data from different unmanned platforms. By constructing a registered trajectory information set using the fused trajectory information of all unmanned platforms, it ensures good spatial consistency of the trajectories of different unmanned platforms. For example, in multi-unmanned vehicle collaborative transportation tasks, this invention can spatially register the trajectory data of each unmanned vehicle, making their movement trajectories more coordinated in space, thereby improving the efficiency and accuracy of collaborative operations. Attached Figure Description
[0049] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation
[0050] To better understand the content of this invention, an embodiment is provided here.
[0051] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.
[0052] In a first aspect, this invention discloses a situational information fusion method for multiple unmanned platforms, comprising:
[0053] S1, acquire geospatial data information and a set of multi-source sensor detection information from the unmanned platform; the set of multi-source sensor detection information includes multi-sensor detection information from each unmanned platform; the multi-sensor detection information includes infrared detection information and radar detection information;
[0054] S2, perform trajectory fusion estimation processing on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform; the trajectory information set includes the trajectory information of each unmanned platform;
[0055] S3, load the set of flight track information into the geospatial data information to obtain the situation information of the unmanned platform; and visualize the situation information.
[0056] The process of performing trajectory fusion estimation on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform includes:
[0057] S21, perform data preprocessing on the multi-source sensor detection information set to obtain a preprocessed information set;
[0058] S22, Perform trajectory estimation processing on the preprocessed information set to obtain the trajectory information set of the unmanned platform.
[0059] The step of preprocessing the multi-source sensor detection information set to obtain a preprocessed information set includes:
[0060] S211, perform data cleaning processing on the multi-source sensor detection information set to obtain the first dataset;
[0061] S212, Smooth the first dataset to obtain the second dataset;
[0062] S213, Perform category discrimination processing on the second dataset to obtain a preprocessed information set;
[0063] The process of performing trajectory estimation on the preprocessed information set to obtain the trajectory information set of the unmanned platform includes:
[0064] S221, The preprocessed information set is processed once to obtain the target motion parameter information set;
[0065] S222, perform secondary processing on the target motion parameter information set to obtain the target trajectory information set;
[0066] S223, the target trajectory information set is processed three times to obtain the flight track information set.
[0067] The step of processing the preprocessed information set to obtain the target motion parameter information set involves performing target detection and motion parameter measurement on the infrared detection information and radar detection information of each target in the multi-sensor detection information of the preprocessed information set to obtain the corresponding target position coordinate information.
[0068] The target detection can be performed using radar signal detection or infrared target detection methods.
[0069] For the measurement of motion parameters, for radar signals, pulse ranging can be used to measure the target distance, and phase angle measurement and amplitude angle measurement can be used to measure the target angle information. The target position coordinates can be obtained using the target distance and target angle information. For infrared information, machine vision, threshold detection or edge detection methods can be used to achieve target detection, and YOLO network can be used to measure the target position coordinate information.
[0070] The secondary processing of the target motion parameter information set to obtain the target trajectory information set includes:
[0071] S2221, Perform trajectory initialization, trajectory extrapolation and trajectory filtering on the target position coordinate information of each unmanned platform in the target motion parameter information set to obtain the trajectory information set of the unmanned platform;
[0072] S2222: Spatial feature registration is performed on the track information set of each unmanned platform to obtain the corresponding registered track information set;
[0073] The process of spatial feature registration of the trajectory information set for each unmanned platform to obtain the corresponding registered trajectory information set includes:
[0074] Clustering is performed on the track information set of each unmanned platform to obtain category information and track information included in each category;
[0075] For each category of track information, a first fusion calculation is performed to obtain the fused track information for each category;
[0076] By utilizing the fused track information from all unmanned platforms, a registration track information set is constructed; the registration track information set includes fused track information of all categories for each unmanned platform.
[0077] The first fusion calculation process is performed on the track information of each category to obtain the fused track information of each category, including:
[0078] For each category of track information, a track matrix is represented; the three row vectors of the track matrix are the sequence of track position coordinates at all times, representing the x-axis, y-axis, and z-axis positions.
[0079] Autocorrelation is calculated on the row vectors of the track matrix to obtain the autocorrelation matrix; each row vector of the autocorrelation matrix is the autocorrelation sequence of the row vectors corresponding to the track matrix.
[0080] Singular value decomposition is performed on the autocorrelation matrix to obtain the singular value matrix;
[0081] Extract the diagonal elements of the singular value matrix to obtain the singular vector;
[0082] Linear fitting is performed on the elements and element index values of the singular vector to obtain the singularity discrimination polynomial;
[0083] Using the singularity discriminant polynomial, the index values of all elements of the singular vector are calculated and processed to obtain the fitted singular vector;
[0084] Based on the autocorrelation matrix, the trajectory matrix is transformed to obtain the transformation matrix;
[0085] The track matrix is smoothed using the transformation matrix and fitted singular vectors of all track information of a category to obtain the fused track matrix; the column vectors of the fused track matrix are the three-dimensional position coordinates of the fused track information at a certain moment.
[0086] The fused track matrix is determined to be the fused track information of the category.
[0087] The singular value decomposition process is calculated using the following expression:
[0088] Y = UAV,
[0089] Where U is the left decomposition matrix, Y is the autocorrelation matrix, A is the singular value matrix, V is the right decomposition matrix, U and V are both orthogonal matrices, and A is the singular value matrix.
[0090] The expression for the transformation process is:
[0091] D = R 1 / 2 FR -1 / 2 ,
[0092] Where F is the track matrix, R is the autocorrelation matrix, and D is the transformation matrix.
[0093] The transformation process, by altering the trajectory matrix F with the square root and inverse of the autocorrelation matrix R, effectively enhances the separability of trajectory features. The autocorrelation matrix R reflects the inherent correlation of the trajectory data, and its square root and inverse operations can re-adjust the feature space of the trajectory data. This transformation makes different trajectory features more clearly separated in the new feature space, facilitating subsequent processing and analysis. In the process of multi-source trajectory data fusion, trajectory data from different unmanned platforms may suffer from feature overlap or confusion. This transformation process effectively solves this problem, making different trajectory features clearer and providing a better foundation for subsequent clustering and fusion calculations.
[0094] The expression for the smoothing process is:
[0095]
[0096] Among them, z ijk Let i be the element in the i-th row and j-th column of the k-th track matrix of a given category. and These are the maximum and minimum track matrices, respectively, f ki Let q be the i-th element of the k-th first feature vector of a category. kiLet be the i-th element of the k-th fitted singular vector for a class, M be the column dimension of the track matrix, N be the number of track matrices for a class, and p ij To merge the elements in the i-th row and j-th column of the track matrix, d ijk It represents the element in the i-th row and j-th column of the k-th transformation matrix of a category.
[0097] The smoothing process, by introducing a Gaussian kernel function, adaptively smooths the trajectory data based on local features. The weights of the Gaussian kernel function vary with the deviation of trajectory data points from their maximum and minimum values, allowing the smoothing process to better adapt to local fluctuations in the trajectory data. For noisy or outlier points in the trajectory data, the smoothing process automatically adjusts the smoothing intensity based on their similarity to surrounding data points, thereby removing noise while preserving the main features of the trajectory. Addressing the noise and discontinuity issues in trajectory data, this smoothing process effectively removes noise and fills in missing data, while avoiding over-smoothing that could lead to feature loss. This adaptive smoothing process better balances the smoothness and feature preservation of trajectory data, improving the accuracy and continuity of the trajectory data.
[0098] This smoothing expression, by incorporating a logarithmic function when calculating weights, allows for dynamic adjustment of the smoothing weights. The introduction of the logarithmic function makes weight adjustment more flexible, enabling dynamic changes in weight magnitude based on the characteristics of the trajectory data. For example, important feature points in the trajectory data will have relatively larger weights, thus better preserving these feature points during the smoothing process. In dynamic trajectory data processing, this smoothing process can dynamically adjust the smoothing weights according to the dynamic changes in the trajectory data, thereby better adapting to sudden changes in the motion state of the unmanned platform. This allows trajectory data processing to more promptly reflect the actual motion of the unmanned platform, improving the dynamic adaptability of trajectory data processing.
[0099] The clustering process can employ methods such as principal component analysis.
[0100] The target trajectory information set is processed three times to obtain the flight path information set, including:
[0101] S2231, Perform track association processing on the target trajectory information set to obtain a target associated track information set; the target associated track information set includes the associated track information set of each unmanned platform;
[0102] S2232, perform a second fusion calculation on the associated track information set of each unmanned platform to obtain the track information of each unmanned platform;
[0103] S2233 utilizes the track information of all unmanned platforms to construct a track information set.
[0104] The second fusion calculation process is performed on the associated trajectory information set of each unmanned platform to obtain the trajectory information of each unmanned platform, including:
[0105] For each unmanned platform, the coordinates of the track points of each associated track information in the associated track information set are taken as the dependent variable, and the acquisition time of the coordinates of the track points of the associated track information in the associated track information set is taken as the independent variable. A polynomial function is fitted to the dependent and independent variables to obtain the track fitting function corresponding to the associated track information.
[0106] Construct a weighted solution model for the trajectory fitting function;
[0107] The expression for the weighting solution model of the trajectory fitting function is:
[0108]
[0109] ω j >0,j=1,2,…,M,
[0110]
[0111] Among them, ga ij Let f be the coordinates of the i-th track point of the j-th associated track information in a set of associated track information for an unmanned platform. j (t i Let be the trajectory fitting function of the j-th associated trajectory information in the set of associated trajectory information of an unmanned platform at the i-th acquisition time t. i The value of is N, where N is the total number of track point coordinates in the associated track information, and M is the total number of associated track information in the associated track information set. Let ω be the second derivative of the trajectory fitting function of the j-th associated trajectory information in the set of associated trajectory information of an unmanned platform, where ∈ is the bias factor to be solved. j Let be the j-th weighting factor to be solved.
[0112] The weighting model of the trajectory fitting function is solved to obtain the values of all weighting factors and bias factors;
[0113] Based on all weighting factors and bias factors, the trajectory function of the unmanned platform is constructed; the expression of the trajectory function is:
[0114]
[0115] Where f0(t) is the trajectory function and t is the time variable;
[0116] By utilizing the trajectory function of the unmanned platform, the acquisition time of the coordinates of the trajectory points associated with the trajectory information is calculated and processed to obtain the trajectory information of the unmanned platform.
[0117] The solution of the weighting model for the trajectory fitting function can be achieved using numerical optimization methods, such as the Gauss-Seidel iterative method.
[0118] The polynomial order of the track set function corresponding to all associated track information is the same.
[0119] The trajectory initialization can be performed using either an intuitive method or a logical method;
[0120] The track extrapolation and track filtering can be performed using the Kalman filtering method.
[0121] The data cleaning process includes filling in missing values and smoothing or deleting outliers; the identification of outliers can be achieved using the Kalman filtering method. The filling value for missing values can be determined by averaging the measurements within a certain sampling interval before and after the missing value.
[0122] The smoothing process involves first identifying the noise data, and then smoothing the noise data based on the data before and after it. The noise data is defined as a value that is less than the sensor's detection sensitivity for the observed data, or greater than the sensor's measurement limit for the observed data.
[0123] The category discrimination process involves judging the attributes of each data point in the second dataset to determine whether the attributes are consistent with preset attributes, and deleting inconsistent data from the second dataset.
[0124] The geospatial data information can be obtained from 3D GIS software or cloud GIS platform, and is 3D spatial mapping data of the geographic space where the unmanned platform is located.
[0125] The step of loading the set of flight track information into the geospatial data information involves adding the flight track information of each unmanned platform in the set of flight track information to the geospatial data information based on its position coordinates, thereby obtaining situational information.
[0126] The visualization of the situation information can be achieved using spatial data visualization software for three-dimensional spatial data, such as virtual reality software or EV-Globe.
[0127] A second aspect of this invention discloses a situational information fusion device for multiple unmanned platforms, the device comprising:
[0128] Memory containing executable program code;
[0129] A processor coupled to the memory;
[0130] The processor calls the executable program code stored in the memory to execute the situation information fusion method for multiple unmanned platforms.
[0131] In a third aspect of this invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, and when the computer instructions are invoked by a computer, they are used to execute the situation information fusion method for multiple unmanned platforms.
[0132] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the situational information fusion method for multiple unmanned platforms.
[0133] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.
Claims
1. A situational information fusion method for multiple unmanned platforms, characterized in that, include: S1, Obtain geospatial data information and a set of multi-source sensor detection information from the unmanned platform; the set of multi-source sensor detection information includes the multi-sensor detection information of each unmanned platform; The multi-sensor detection information includes infrared detection information and radar detection information; S2, perform trajectory fusion estimation processing on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform; the trajectory information set includes the trajectory information of each unmanned platform; S3, load the set of flight track information into the geospatial data information to obtain the situation information of the unmanned platform; and visualize the situation information.
2. The situational information fusion method for multiple unmanned platforms as described in claim 1, characterized in that, The process of performing trajectory fusion estimation on the multi-source sensor detection information set to obtain the trajectory information set of the unmanned platform includes: S21, perform data preprocessing on the multi-source sensor detection information set to obtain a preprocessed information set; S22, perform trajectory estimation processing on the preprocessed information set to obtain the trajectory information set of the unmanned platform.
3. The situational information fusion method for multiple unmanned platforms as described in claim 1, characterized in that, The step of preprocessing the multi-source sensor detection information set to obtain a preprocessed information set includes: S211, perform data cleaning processing on the multi-source sensor detection information set to obtain the first dataset; S212, Smooth the first dataset to obtain the second dataset; S213, perform category discrimination processing on the second dataset to obtain a preprocessed information set.
4. The situational information fusion method for multiple unmanned platforms as described in claim 1, characterized in that, The process of performing trajectory estimation on the preprocessed information set to obtain the trajectory information set of the unmanned platform includes: S221, The preprocessed information set is processed once to obtain the target motion parameter information set; S222, perform secondary processing on the target motion parameter information set to obtain the target trajectory information set; S223, the target trajectory information set is processed three times to obtain the flight track information set.
5. The situational information fusion method for multiple unmanned platforms as described in claim 4, characterized in that, The secondary processing of the target motion parameter information set to obtain the target trajectory information set includes: S2221, Perform trajectory initialization, trajectory extrapolation and trajectory filtering on the target position coordinate information of each unmanned platform in the target motion parameter information set to obtain the trajectory information set of the unmanned platform; S2222: Spatial feature registration is performed on the track information set of each unmanned platform to obtain the corresponding registered track information set.
6. The situational information fusion method for multiple unmanned platforms as described in claim 5, characterized in that, The process of spatial feature registration of the trajectory information set for each unmanned platform to obtain the corresponding registered trajectory information set includes: Clustering is performed on the track information set of each unmanned platform to obtain category information and track information included in each category; For each category of track information, a first fusion calculation is performed to obtain the fused track information for each category; By utilizing the fused track information from all unmanned platforms, a registration track information set is constructed; the registration track information set includes fused track information of all categories for each unmanned platform.
7. The situational information fusion method for multiple unmanned platforms as described in claim 6, characterized in that, The first fusion calculation process is performed on the track information of each category to obtain the fused track information of each category, including: For each category of track information, a track matrix is represented; the three row vectors of the track matrix are the sequence of track position coordinates at all times, representing the x-axis, y-axis, and z-axis positions. Autocorrelation is calculated on the row vectors of the track matrix to obtain the autocorrelation matrix; each row vector of the autocorrelation matrix is the autocorrelation sequence of the row vectors corresponding to the track matrix. Singular value decomposition is performed on the autocorrelation matrix to obtain the singular value matrix; Extract the diagonal elements of the singular value matrix to obtain the singular vector; Linear fitting is performed on the elements and element index values of the singular vector to obtain the singularity discrimination polynomial; Using the singularity discriminant polynomial, the index values of all elements of the singular vector are calculated and processed to obtain the fitted singular vector; Based on the autocorrelation matrix, the trajectory matrix is transformed to obtain the transformation matrix; The track matrix is smoothed using the transformation matrix and fitted singular vectors of all track information of a category to obtain the fused track matrix; the column vectors of the fused track matrix are the three-dimensional position coordinates of the fused track information at a certain moment. The fused track matrix is determined to be the fused track information of the category.
8. A situational information fusion device for multiple unmanned platforms, 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 situation information fusion method for multiple unmanned platforms as described in any one of claims 1 to 7.
9. 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 situation information fusion method for multiple unmanned platforms as described in any one of claims 1 to 7.
10. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the situational information fusion method for multiple unmanned platforms as described in any one of claims 1 to 7.