A Beidou airborne data fusion method, system and device
By constructing a multi-type filtering fusion model and integrating Beidou navigation, inertial navigation, airborne meteorological and ground auxiliary data, the problem of low accuracy of traditional navigation data fusion technology in nonlinear and non-Gaussian distributed data processing is solved, and navigation parameters are improved accuracy and stability.
Patent Information
- Application Number
- CN202411204240.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-30
AI Technical Summary
Traditional navigation data fusion technology has high accuracy in model dependence, Kalman filtering algorithm has limited effect in processing nonlinear and non-Gaussian distributed data, and sensor calibration is susceptible to single sensor anomalies, resulting in low accuracy of navigation parameters.
By obtaining Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data, a multi-type filter fusion model is built, various types of data are fused to generate filter fusion data, and navigation calculations are carried out to obtain navigation parameters.
It significantly improves the accuracy and stability of navigation parameters after navigation data fusion, reduces errors, enhances the adaptability of navigation systems in complex environments, and improves the overall performance of navigation systems.
Smart Images

Figure CN118746304B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a Beidou airborne data fusion method, system and device. Background Art
[0002] With the development of computer technology, navigation data fusion technology has emerged. This technology is the process of comprehensive processing and analysis of data from multiple navigation sensors and systems (such as GPS, inertial navigation system, Beidou satellite navigation system, etc.). By fusing these multi-source data, positioning accuracy, reliability and robustness can be improved, the impact of single sensor data errors can be reduced, and more stable and accurate navigation information can be provided to meet navigation needs in various complex environments.
[0003] In traditional technology, navigation data fusion technology is mainly achieved through Kalman filtering algorithm and sensor calibration, which is highly dependent on model accuracy. When the model is inaccurate or the environment is complex, the filter performance may degrade. When processing nonlinear and non-Gaussian distribution data, the effect of the Kalman filtering algorithm is limited, and sensor calibration is easily affected by the anomalies of single sensor data. This results in low accuracy of the corresponding navigation parameters generated after navigation data fusion. Summary of the invention
[0004] Based on this, it is necessary to provide a Beidou airborne data fusion method, system, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of generating corresponding navigation parameters after navigation data fusion in order to solve the above technical problems.
[0005] In a first aspect, the present application provides a Beidou airborne data fusion method, comprising:
[0006] Acquire Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to an airborne navigation system; construct a multi-type filter fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; input the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; fuse each airborne filter fusion data to obtain the airborne fused navigation data corresponding to the airborne navigation system; perform navigation solution on the airborne fused data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0007] In the second aspect, the present application also provides a Beidou airborne data fusion system. The system includes: a Beidou navigation terminal, an inertial navigation device, a meteorological collection terminal, a ground data transmission terminal and a server; the Beidou navigation terminal is used to obtain Beidou navigation data and send the Beidou navigation data to the server; the inertial navigation device is used to obtain inertial navigation data and send the inertial navigation data to the server; the meteorological collection terminal is used to obtain airborne meteorological data and send the airborne meteorological data to the server; the ground data transmission terminal is used to send the collected ground auxiliary data to the server; the server is used to collect the ground auxiliary data according to the Beidou navigation data, the ground auxiliary data and the ground auxiliary data. The inertial navigation data, the airborne meteorological data and the ground auxiliary data are used to construct a multi-type filtering fusion model corresponding to the airborne navigation system; the server is also used to input the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filtering fusion model to obtain each airborne filtering fusion data; the server is also used to fuse each airborne filtering fusion data to obtain the airborne fused navigation data corresponding to the airborne navigation system; the server is also used to perform navigation solution on the airborne fused data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0008] In the third aspect, the present application also provides a Beidou airborne data fusion device. The device includes: a navigation data acquisition module, which is used to acquire Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; a fusion model construction module, which is used to construct a multi-type filter fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; a fusion data acquisition module, which is used to input the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; a navigation data acquisition module, which is used to fuse each airborne filter fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system; a navigation parameter acquisition module, which is used to perform navigation solution on the airborne fusion data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0009] In a fourth aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program. When the processor executes the computer program, the following steps are implemented: obtaining Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; constructing a multi-type filtering fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; inputting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filtering fusion model to obtain each airborne filtering fusion data; fusing each of the airborne filtering fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system; performing navigation solution on the airborne fusion data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0010] In a fifth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the following steps: obtaining Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to an airborne navigation system; constructing a multi-type filtering fusion model corresponding to the airborne navigation system based on the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; inputting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filtering fusion model to obtain each airborne filtering fusion data; fusing each airborne filtering fusion data to obtain the airborne fused navigation data corresponding to the airborne navigation system; performing navigation solution on the airborne fused data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0011] In a sixth aspect, the present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the following steps: obtaining Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to an airborne navigation system; constructing a multi-type filtering fusion model corresponding to the airborne navigation system based on the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; inputting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filtering fusion model to obtain each airborne filtering fusion data; fusing each airborne filtering fusion data to obtain the airborne fused navigation data corresponding to the airborne navigation system; performing navigation solution on the airborne fused data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0012] The above-mentioned Beidou airborne data fusion method, system, device, computer equipment, storage medium and computer program product obtain Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; construct a multi-type filter fusion model corresponding to the airborne navigation system according to the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data; input the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; fuse each airborne filter fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system; perform navigation solution on the airborne fusion data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0013] By acquiring Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data, and constructing a multi-type filter fusion model, various types of data are input into the model to generate multiple airborne filter fusion data, and then multiple airborne filter fusion data are fused to obtain airborne fusion navigation data, and finally navigation solution is performed to obtain accurate airborne navigation parameters. This method can significantly improve the accuracy, stability and reliability of the corresponding navigation parameters generated after navigation data fusion, and can provide more accurate navigation information in complex environments, reduce errors, and improve the adaptability of the navigation system under various conditions, thereby improving the safety and efficiency of equipment movement. In addition, by comprehensively utilizing multiple data sources, the uncertainty and errors that may be caused by a single data source can be effectively reduced, data complementarity and optimization can be achieved, and the overall performance of the navigation system can be further enhanced. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0015] Figure 1 An application environment diagram of a Beidou airborne data fusion method in an embodiment;
[0016] Figure 2 A schematic diagram of a flow chart of a Beidou airborne data fusion method in an embodiment;
[0017] Figure 3 It is a flowchart of a method for obtaining airborne fused navigation data in one embodiment;
[0018] Figure 4 A schematic flow chart of a method for obtaining airborne fused navigation data in another embodiment;
[0019] Figure 5 It is a flowchart of a method for obtaining initial airborne fused navigation data in one embodiment;
[0020] Figure 6 It is a flowchart of a method for obtaining airborne fused navigation data in another embodiment;
[0021] Figure 7 It is a flowchart of a method for obtaining airborne navigation parameters in one embodiment;
[0022] Figure 8 A schematic diagram of a flow chart of a method for obtaining coordinate position information in one embodiment;
[0023] Fig. 9 A schematic diagram of a flow chart of a method for obtaining airborne filtering and fusion data in one embodiment;
[0024] Fig.10 It is a structural block diagram of a Beidou airborne data fusion device in an embodiment;
[0025] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0026] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0027] The Beidou airborne data fusion method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; according to the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data, a multi-type filter fusion model corresponding to the airborne navigation system is constructed; the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data are input into the multi-type filter fusion model to obtain each airborne filter fusion data; the airborne filter fusion data is fused to obtain the airborne fusion navigation data corresponding to the airborne navigation system; the airborne fusion data is navigated and solved to obtain the airborne navigation parameters corresponding to the airborne navigation system. Among them, the server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.
[0028] In an exemplary embodiment, Figure 2 As shown, a Beidou airborne data fusion method is provided, and the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 202 to 210. Among them:
[0029] Step 202, obtaining Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system.
[0030] Among them, Beidou navigation data can be the positioning, navigation and timing information provided by the BeiDou Navigation Satellite System (BDS). These data include geographical location coordinates, time synchronization signals and navigation information around the world.
[0031] Among them, the inertial navigation system can be a navigation system independent of external references. It uses an inertial measurement unit (IMU) to measure the acceleration and angular velocity of the movement of aircraft, ships, vehicles, etc., and then estimates the position, speed and direction by integrating these measurements.
[0032] Airborne meteorological data refers to data collected by various sensors on aircraft (such as airplanes or drones) for obtaining atmospheric and weather information. These data are used by pilots, flight control systems or ground meteorological service departments to make flight decisions, navigation plans and weather forecasts.
[0033] Among them, ground auxiliary data can be data used to assist navigation, positioning and map display, and these data are provided by ground facilities or satellite ground stations.
[0034] Specifically, it is necessary to obtain the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data required by the airborne navigation system. First, the corresponding sensors and receiver equipment need to be installed. For Beidou navigation data, a Beidou receiver needs to be installed to obtain positioning and time information by receiving signals from Beidou satellites to obtain Beidou navigation data. For inertial navigation data, an inertial measurement unit (IMU) sensor needs to be installed to measure the acceleration and angular velocity of the aircraft to obtain inertial navigation data. Airborne meteorological data can be obtained by installing meteorological radars, atmospheric pressure sensors, meteorological satellite receivers and other equipment on the aircraft to obtain atmospheric and weather information and obtain airborne meteorological data. Ground auxiliary data needs to be obtained from ground servers or ground sensors through communication networks or dedicated interfaces, such as differential GPS data, landmark data, terrain and elevation data, traffic data, etc.
[0035] Step 204, constructing a multi-type filtering fusion model corresponding to the airborne navigation system according to Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data.
[0036] Among them, the multi-type filter fusion model can be a fusion model that processes and fuses data from different sensors by integrating multiple filtering methods (such as particle filter algorithm, unscented Kalman filter algorithm, ensemble learning algorithm, adaptive information fusion algorithm, and Bayesian filter algorithm, etc.). This fusion model can improve the accuracy and robustness of the navigation system, ensuring reliable and accurate position, velocity, and attitude information in complex and changing flight environments. Through the collaborative work of multiple types of filters, the system can more effectively deal with sensor noise and errors and improve overall navigation performance.
[0037] Specifically, the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data are respectively input into the data feature extraction model, and the features of the above data are extracted through the neural network of the data feature extraction model to obtain Beidou navigation feature data, inertial navigation feature data, airborne meteorological feature data and ground auxiliary feature data; further, according to the Beidou navigation feature data, inertial navigation feature data, airborne meteorological feature data and ground auxiliary feature data, different types of filtering models or filtering algorithms are selected to construct a multi-type filtering fusion model corresponding to the airborne navigation system, for example, a particle filtering algorithm, an unscented Kalman filtering algorithm, an ensemble learning algorithm, an adaptive information fusion algorithm and a Bayesian filtering algorithm are selected to construct a multi-type filtering fusion model.
[0038] Step 206, inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into a multi-type filter fusion model to obtain each airborne filter fusion data.
[0039] Among them, airborne filtered fusion data refers to the data processed by filtering fusion technology in the airborne system. These data are obtained by fusing, filtering and processing information from different sensors or data sources.
[0040] Specifically, feature extraction and state estimation are performed on Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data respectively. For example, the position and speed of Beidou navigation data are estimated, the attitude and acceleration of inertial navigation data are estimated, the wind speed and air pressure of airborne meteorological data are estimated, and the landmarks and roads of ground auxiliary data are identified. Subsequently, the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data after state estimation are fused respectively by using various filtering algorithms in the multi-type filtering fusion model, such as particle filtering algorithm, unscented Kalman filtering algorithm, integrated learning algorithm, adaptive information fusion algorithm and Bayesian filtering algorithm. In the fusion process, the data are weightedly fused according to the weight and confidence of each data source to obtain the optimal state estimation or prediction. Finally, the fused airborne filtering fusion data is output, including the fused position, speed, attitude and other information, as well as the corresponding uncertainty or confidence index.
[0041] Step 208, fusing the airborne filtered fusion data to obtain airborne fusion navigation data corresponding to the airborne navigation system;
[0042] Among them, airborne fused navigation data refers to the data used for navigation purposes obtained by fusing the analysis results from multiple different filtering algorithms or filtering models. These data contain estimated or predicted information on the aircraft's position, speed, attitude and other states, and are the output of the airborne navigation system.
[0043] Specifically, the first step is to determine the weight and confidence of each airborne filter fusion data, and weight each airborne filter fusion data according to the reliability and importance of each airborne filter fusion data. The second step is to use each fusion algorithm under the hybrid strategy fusion strategy to fuse the information of each airborne filter fusion data. During the fusion process, according to the weight and confidence of each airborne filter fusion data, a layer-by-layer fusion algorithm is used to perform weighted fusion on the data to obtain the optimal airborne fusion navigation data. Finally, the fused airborne fusion navigation data is output, including the fused position, velocity, attitude and other information, as well as the corresponding uncertainty or confidence indicators.
[0044] Step 210, performing navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system.
[0045] Navigation solution refers to the process of calculating the position, speed, attitude and other navigation status of aircraft, ships, vehicles or other mobile platforms through mathematical and physical models using information provided by various sensors or data sources. The purpose of navigation solution is to determine the position and direction of the mobile platform in space in order to achieve effective navigation and positioning.
[0046] Specifically, it is necessary to first extract navigation status information such as position, speed, and attitude from the fused data. Then, the information is solved using mathematical models and navigation algorithms, where the solving process involves operations such as integration, differentiation, or filtering to obtain a more accurate navigation status estimate. Finally, the solved airborne navigation parameters corresponding to the airborne navigation system are used by the airborne navigation system to achieve effective navigation and positioning of aircraft or other mobile platforms.
[0047] In the above-mentioned Beidou airborne data fusion method, Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system are obtained; a multi-type filter fusion model corresponding to the airborne navigation system is constructed according to the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data; the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data are input into the multi-type filter fusion model to obtain each airborne filter fusion data; the airborne filter fusion data are fused to obtain the airborne fused navigation data corresponding to the airborne navigation system; navigation solution is performed on the airborne fused data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0048] By acquiring Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data, and constructing a multi-type filter fusion model, various types of data are input into the model to generate multiple airborne filter fusion data, and then multiple airborne filter fusion data are fused to obtain airborne fusion navigation data, and finally navigation solution is performed to obtain accurate airborne navigation parameters. This method can significantly improve the accuracy, stability and reliability of the corresponding navigation parameters generated after navigation data fusion, and can provide more accurate navigation information in complex environments, reduce errors, and improve the adaptability of the navigation system under various conditions, thereby improving the safety and efficiency of equipment movement. In addition, by comprehensively utilizing multiple data sources, the uncertainty and errors that may be caused by a single data source can be effectively reduced, data complementarity and optimization can be achieved, and the overall performance of the navigation system can be further enhanced.
[0049] In an exemplary embodiment, Figure 3 As shown, the airborne filtering fusion data of each airborne navigation system is fused to obtain the airborne fusion navigation data corresponding to the airborne navigation system, including:
[0050] Step 302: Determine navigation data fusion strategy information based on feature information of each airborne filtered fusion data.
[0051] The navigation data fusion strategy information may refer to relevant information used to guide the fusion algorithm on how to perform data fusion during the navigation data fusion process, and such information generally includes weight allocation, confidence assessment, data quality assessment, fusion rules, and the like.
[0052] Specifically, neural networks are used to extract features from each airborne filter fusion data to obtain feature information of each airborne filter fusion data, and further feature information of each airborne filter fusion data is analyzed, including data accuracy, reliability, dynamics, etc. At the same time, according to the feature information of each airborne filter fusion data, the weight distribution of each airborne filter fusion data in the fusion process is determined to reflect its relative contribution; at the same time, confidence assessment is performed on each airborne filter fusion data to evaluate the data credibility and reliability of each airborne filter fusion data; at the same time, the quality assessment of the data provided by each airborne filter fusion data is performed, including data accuracy, stability, etc. Finally, according to the results of weight allocation, confidence assessment and data quality assessment, navigation data fusion strategy information is formulated, including fusion rules, selection of filtering algorithms, etc.
[0053] Step 304: fuse the airborne filtered fusion data according to the navigation data fusion strategy information to obtain airborne fusion navigation data.
[0054] Specifically, according to the guidance in the navigation data fusion strategy information, after selecting several navigation data fusion algorithms, each airborne filter fusion data is input into the navigation data fusion algorithm in turn, and each navigation data fusion algorithm executes the fusion of each airborne filter fusion data to obtain the initial airborne fusion navigation data corresponding to each navigation fusion algorithm; if each initial airborne fusion navigation data can pass the verification, one of the above-mentioned navigation data fusion algorithms is executed again to obtain the airborne fusion navigation data for use by the navigation system.
[0055] In this embodiment, by determining the navigation data fusion strategy information according to the characteristic information of each airborne filtered fusion data, the data fusion process can be optimized, so that the system can more effectively utilize the information provided by each data source and improve the accuracy and robustness of the navigation solution. Subsequently, according to the determined navigation data fusion strategy information, each airborne filtered fusion data is fused to obtain airborne fused navigation data. Such a method can effectively and comprehensively utilize multi-source data, reduce data errors, improve the performance and reliability of the navigation system, and provide pilots with more accurate navigation information, thereby promoting flight safety and efficiency.
[0056] In an exemplary embodiment, Figure 4 As shown, according to the navigation data fusion strategy information, the airborne filter fusion data are fused to obtain the airborne fusion navigation data including:
[0057] Step 402: when the navigation data fusion strategy information indicates that the fusion mode is a hybrid strategy fusion, determine the navigation data fusion algorithms corresponding to the airborne navigation system.
[0058] Among them, the hybrid strategy fusion can be to use a variety of different navigation data fusion algorithms in a random order during the data fusion process to combine different airborne filter fusion data or information of processing steps. In the hybrid strategy fusion, multiple fusion rules can also be mixed to process data according to actual conditions to achieve the best fusion effect.
[0059] The navigation data fusion algorithm may be an algorithm for integrating and fusing filtering fusion data from a plurality of different airborne devices.
[0060] Specifically, when the navigation data fusion strategy information characterization fusion mode is a hybrid strategy fusion, the characteristic information of each airborne filter fusion data is comprehensively analyzed, including the quality, stability, accuracy and other aspects of the data source. Secondly, according to the analysis results, select and adjust the appropriate navigation data fusion algorithms, such as particle filter algorithm, unscented Kalman filter algorithm, ensemble learning algorithm, adaptive information fusion algorithm and Bayesian filter algorithm. Then, according to the actual situation, adjust the parameters of the selected navigation data fusion algorithm to adapt to different working environments and application scenarios. Finally, integrate the selected and adjusted navigation data fusion algorithms into the airborne navigation system to ensure that the system can fuse and process information from different data sources on time and as needed.
[0061] Step 404: According to the navigation data fusion algorithms, the airborne filtered fusion data are fused respectively to obtain initial airborne fusion navigation data.
[0062] Specifically, for each navigation data fusion algorithm, the corresponding airborne filtered fusion data is input into the corresponding algorithm for fusion processing. According to the principles and requirements of the algorithm, data fusion calculation and optimization are performed, and information from different data sources is merged and integrated. For each navigation data fusion algorithm, the corresponding initial airborne fusion navigation data is obtained, which contains the navigation information after comprehensive processing.
[0063] Step 406, when the difference between the fused navigation parameter curve of any initial airborne fused navigation data and the current navigation parameter curve is greater than a threshold, adjust the parameters of each navigation data fusion algorithm, and return to execute according to each navigation data fusion algorithm, respectively fuse each airborne filtered fusion data to obtain each initial airborne fused navigation data, until the difference between the fused navigation parameter curve and the current navigation parameter curve is less than a threshold, and obtain the airborne fused navigation data according to each initial airborne fused navigation data.
[0064] Specifically, the current navigation parameter curve is compared with the fused navigation parameter curve of the initial airborne fused navigation data. When the difference between the fused navigation parameter curve of any initial airborne fused navigation data and the current navigation parameter curve is greater than a threshold value, the parameter adjustment step is performed to adjust the parameters of each navigation data fusion algorithm. In the parameter adjustment step, the parameters of each navigation data fusion algorithm are adjusted accordingly according to the difference. Each navigation data fusion algorithm is re-executed to fuse each airborne filtered fusion data, and each navigation data fusion algorithm after adjusting the parameters is applied to the fusion process to obtain new initial airborne fused navigation data. The difference between the new fused navigation parameter curve and the current navigation parameter curve is compared again. If the difference is still greater than the threshold value, the parameter adjustment and fusion steps are continued. Until the difference is less than the threshold value, it can be considered that the fusion result is close enough to the actual navigation parameter, and then the airborne fused navigation data is obtained according to the final initial airborne fused navigation data.
[0065] In this embodiment, the accuracy and stability of the fusion result are improved by continuously adjusting the parameters of the navigation data fusion algorithm. First, each initial airborne fusion navigation data is obtained by fusing the airborne filtered fusion data respectively, so that information from different data sources can be integrated. Then, when the difference between the fusion navigation parameters of any initial airborne fusion navigation data and the current navigation parameter curve is greater than a threshold, the system will automatically adjust the parameters of the navigation data fusion algorithm. This dynamic adjustment can ensure that the fusion navigation parameters can be as close to the actual navigation parameters as possible, thereby improving the accuracy and robustness of data fusion. By repeatedly executing the fusion algorithm until the difference is less than the threshold, the airborne fusion navigation data finally obtained can more accurately reflect the flight status and provide more reliable support for flight control and navigation operations.
[0066] In one embodiment, Figure 5 As shown, each airborne filter fusion data includes particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data; in the case where one of the navigation data fusion algorithms is a layer-by-layer fusion algorithm; according to each navigation data fusion algorithm, each airborne filter fusion data is fused respectively to obtain each initial airborne fusion navigation data, including:
[0067] Step 502, according to the layer-by-layer fusion algorithm, select several target filtering algorithms from the various fusion filtering algorithms of the multi-type filtering fusion model; the target filtering algorithms include Gaussian process fusion algorithm, sparse Bayesian learning algorithm, deep reinforcement learning fusion algorithm and integrated Kalman filtering algorithm.
[0068] Among them, the layer-by-layer fusion algorithm is a data fusion method, which is usually used for information fusion in a multi-level structure. In the layer-by-layer fusion algorithm, the data fusion process is divided into multiple levels, each level performs certain processing and fusion on the data, and then passes the results to the next level for further processing and fusion. This hierarchical fusion method can help the system better manage and process large amounts of data, and can also gradually extract and integrate information from the data, thereby improving the overall fusion effect. The layer-by-layer fusion algorithm usually includes multiple steps, each step is performed on the basis of the previous step.
[0069] The target filtering algorithm may be a fusion filtering algorithm selected to perform fusion calculations.
[0070] Among them, the Gaussian process fusion algorithm can be a probability-based machine learning method for dealing with uncertainty and nonlinear data modeling. It is an algorithm that fuses the prediction results of the Gaussian process with information from multiple data sources.
[0071] Among them, the sparse Bayesian learning algorithm can be a Bayesian method for processing high-dimensional data, and its main goal is to achieve sparsity of model parameters in the learning process to achieve an algorithm with fusion functions. In sparse Bayesian learning, it is usually assumed that the prior distribution of model parameters is a sparse distribution, which means that most parameters are zero or close to zero, and only a small number of parameters play an important role in data fitting.
[0072] Among them, the deep reinforcement learning fusion algorithm can be a method that combines deep learning and reinforcement learning to solve complex decision-making problems. In this algorithm, deep learning is used to learn the state representation or decision-making strategy of the environment, while reinforcement learning is used to optimize these learned strategies.
[0073] Among them, the integrated Kalman filter algorithm can be a technology that fuses multiple Kalman filters together to improve the accuracy and robustness of state estimation. By running multiple Kalman filters at the same time, each filter can process different data sources or model assumptions, and then the estimation results of each filter are weighted and fused to obtain a more accurate and stable state estimate. This method is widely used in multi-sensor data fusion, navigation systems, signal processing and other fields, and can effectively deal with model errors and measurement noise that may occur in a single filter.
[0074] Specifically, the process of implementing the layer-by-layer fusion algorithm involves selecting target filter algorithms from various fusion filter algorithms of multi-type filter fusion models and combining them together to achieve higher-level state estimation or prediction. First, according to the task requirements and system characteristics given in the layer-by-layer fusion algorithm, several target filter algorithms are selected from the given fusion filter algorithms, including Gaussian process fusion algorithm, sparse Bayesian learning algorithm, deep reinforcement learning fusion algorithm and integrated Kalman filter algorithm. Then, the selected target filter algorithms are combined together according to the guidance of the layer-by-layer fusion algorithm, and parallel, cascade, alternating and other strategies may be adopted to determine the fusion method according to their respective advantages and applicable scenarios.
[0075] Step 504: According to the fusion step information of the layer-by-layer fusion algorithm, a Gaussian process fusion algorithm is used to fuse the particle filter fusion data and the unscented Kalman filter fusion data to obtain the first layer of fusion data.
[0076] Among them, the particle filter fusion data can be the output data obtained by inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the particle filter algorithm for calculation.
[0077] The unscented Kalman filter fusion data may be output data obtained by inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the unscented Kalman filter algorithm for calculation.
[0078] The first layer of fusion data may be the result of the first fusion step in the layer-by-layer fusion algorithm.
[0079] Specifically, according to the fusion step information of the layer-by-layer fusion algorithm, after defining the kernel function and hyperparameters in the Gaussian process fusion algorithm, the particle filter fusion data and the unscented Kalman filter fusion data are modeled through the Gaussian process fusion algorithm, and their covariance matrix is calculated. The model parameters are solved using the maximum likelihood estimation method to achieve the fusion of the particle filter fusion data and the unscented Kalman filter fusion data to obtain the first layer of fusion data.
[0080] Step 506: According to the fusion step information of the layer-by-layer fusion algorithm, a sparse Bayesian learning algorithm is used to fuse the first layer fusion data and the ensemble learning fusion data to obtain the second layer fusion data.
[0081] The second layer of fusion data may be the result of the second step fusion in the layer-by-layer fusion algorithm.
[0082] Among them, the integrated learning fusion data can be the output data obtained by inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the integrated learning algorithm for calculation.
[0083] Specifically, according to the fusion step information of the layer-by-layer fusion algorithm, the first-layer fusion data and the ensemble learning fusion data are modeled through the sparse Bayesian model, the posterior probability distribution of the data is calculated using the Bayesian reasoning method, and the most important features and parameters are selected through sparsity constraints. Finally, the first-layer fusion data and the ensemble learning fusion data are combined through the sparse Bayesian learning algorithm to obtain the second-layer fusion data.
[0084] Step 508: According to the fusion step information of the layer-by-layer fusion algorithm, a deep reinforcement learning fusion algorithm is used to fuse the second layer fusion data and the adaptive information fusion data to obtain the third layer fusion data.
[0085] The third layer fusion data may be the result of the third step fusion in the layer-by-layer fusion algorithm.
[0086] Among them, the adaptive information fusion data can be the output data obtained by inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the adaptive information fusion algorithm for calculation.
[0087] Specifically, according to the fusion step information of the layer-by-layer fusion algorithm, a deep reinforcement learning fusion algorithm is used to build a deep reinforcement learning model, with the second-layer fusion data and adaptive information fusion data as input, and the corresponding reward function and strategy network are designed. The model parameters are optimized through multiple training iterations. The model selects the optimal action through the strategy at each step, updates the strategy and value function, and gradually improves the fusion effect to obtain the third-layer fusion data.
[0088] Step 510: According to the fusion step information of the layer-by-layer fusion algorithm, the third layer fusion data and the Bayesian filter fusion data are fused using the integrated Kalman filter algorithm to obtain initial airborne fusion navigation data.
[0089] Among them, the Bayesian filtering fusion data can be the output data obtained by inputting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the Bayesian filtering algorithm for calculation.
[0090] Specifically, the third-layer fusion data and the Bayesian filter fusion data are predicted and updated respectively by integrating the Kalman filter algorithm, and the Bayesian filter fusion data is used as a priori information, combined with a variety of different Kalman filter prediction and observation update mechanisms, to gradually correct the state estimation of the fusion of the third-layer fusion data and the Bayesian filter fusion data. Finally, the initial airborne fusion navigation data is obtained by integrating the advantages of the two filter data through the Kalman filter algorithm.
[0091] In this embodiment, through the steps of the above layer-by-layer fusion algorithm, different target filtering algorithms in the multi-type filtering fusion model are utilized, including Gaussian process fusion algorithm, sparse Bayesian learning algorithm, deep reinforcement learning fusion algorithm and integrated Kalman filtering algorithm; this series of fusion steps fully utilizes the advantages of different filtering algorithms, effectively improves the accuracy and reliability of data fusion, and provides more accurate navigation data support for the airborne navigation system, thereby improving the stability and accuracy of navigation.
[0092] In an exemplary embodiment, Figure 6 As shown, according to each initial airborne fusion navigation data, the airborne fusion navigation data is obtained, including:
[0093] Step 602: Fuse any two selected initial airborne fused navigation data to obtain intermediate fused airborne navigation data.
[0094] Specifically, two sets of initial airborne fusion navigation data are prepared as input, and then a suitable navigation data fusion algorithm is selected, such as Gaussian process fusion algorithm, sparse Bayesian learning algorithm, deep reinforcement learning fusion algorithm and integrated Kalman filter algorithm, to fuse the two sets of initial airborne fusion navigation data to obtain intermediate fusion airborne navigation data. In this process, the weight or interpolation parameter can be adjusted according to different situations to achieve the best fusion effect.
[0095] Step 604, performing error correction on the intermediate fused airborne navigation data according to the ground auxiliary data to obtain intermediate corrected airborne navigation data.
[0096] Specifically, the ground auxiliary data and the intermediate fused airborne navigation data are matched, an error model is established, and a filtering algorithm is used to estimate the error, and the intermediate fused airborne navigation data is corrected to finally obtain intermediate corrected airborne navigation data with higher accuracy.
[0097] Step 606, taking the intermediate corrected airborne navigation data as one of the selected initial airborne fusion navigation data, selecting any initial airborne fusion navigation data from the unselected initial airborne fusion navigation data, returning to execute the step of fusing any two selected initial airborne fusion navigation data to obtain the intermediate fusion airborne navigation data, until all the initial airborne fusion navigation data have been selected at least once to obtain the airborne fusion navigation data.
[0098] Specifically, the intermediate corrected airborne navigation data is set as one of the selected initial airborne fusion navigation data in the initial airborne fusion navigation data set, denoted as D0, and the other initial airborne fusion navigation data sets are stored in a list, denoted as [D1, D2, ..., Dn]. D0 is selected as the first selected initial airborne fusion navigation data. A data set is randomly selected from the unselected initial airborne fusion navigation data [D1, D2, ..., Dn], denoted as Di. Data fusion is performed on D0 and Di to obtain new intermediate fusion airborne navigation data, denoted as D'. The new intermediate fusion airborne navigation data D' is updated to the new D0. Di is removed from the list of unselected initial airborne fusion navigation data. Di is marked as selected. A data set is randomly selected from the unselected initial airborne fusion navigation data [D1, D2, ..., Dn] and subsequent steps are repeated until all initial airborne fusion navigation data [D1, D2, ..., Dn] have been selected at least once. When all the initial airborne fused navigation data are selected and fused at least once, the final airborne fused navigation data is obtained, which is recorded as D_final.
[0099] In this embodiment, by sequentially fusing any two initial airborne fused navigation data and performing error correction based on ground auxiliary data, the corrected data is then used as new initial data to continue fusing and correcting with the unselected data, thereby gradually improving the accuracy and reliability of the navigation data. This process ensures that each initial navigation data is selected and fused at least once, so that the final airborne fused navigation data has higher accuracy and consistency, significantly improving the performance of the navigation system.
[0100] In an exemplary embodiment, Figure 7 As shown, the airborne fusion data is subjected to navigation solution to obtain the airborne navigation parameters corresponding to the airborne navigation system, including:
[0101] Step 702, determine the position calculation algorithm, speed calculation algorithm and heading calculation algorithm corresponding to the airborne fusion line data.
[0102] Among them, the position solution algorithm refers to the mathematical method and calculation process used to calculate the precise position of an object in three-dimensional space. This algorithm is often used in navigation systems, geographic information systems (GIS), self-driving cars, drones, and other applications that require precise position measurement. The core of the position solution algorithm is to determine the coordinates of an object by processing and analyzing sensor data (such as GPS, INS, radar, lidar, etc.).
[0103] Among them, the velocity solution algorithm is a mathematical method and calculation process used to calculate the speed of an object in three-dimensional space. These algorithms process sensor data (such as GPS, IMU, radar, etc.) to solve the object's speed information. Velocity solution plays a key role in navigation, autonomous driving, aircraft control and other fields, and can provide accurate motion state estimation.
[0104] Among them, heading solution algorithms are mathematical methods and calculation processes used to determine the direction of an object (such as a ship, aircraft, vehicle, etc.) in three-dimensional space. These algorithms calculate the heading of an object (i.e., the direction angle relative to the north) by processing and fusing various sensor data (such as magnetic compass, gyroscope, GPS, etc.). Heading solution is crucial in the fields of navigation, autonomous driving, aircraft control, etc., and can provide accurate direction information.
[0105] Specifically, a position solution algorithm is constructed by using triangulation or satellite positioning technology through the position information fused into the line data. Similarly, a velocity solution algorithm is constructed by combining the data of the acceleration sensor and the velocity sensor with the rate of change of the position fused into the line data over time. Finally, a heading solution algorithm is constructed using the direction data provided by the geomagnetic sensor and the gyroscope.
[0106] Step 704, solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in the fixed coordinate system.
[0107] Specifically, the airborne fused data is input into the position solution algorithm, and the triangulation calculation or satellite positioning calculation is implemented through the position solution algorithm to calculate the coordinate position information relative to the fixed coordinate system. This process includes data filtering, error correction and multi-point position fusion, and finally the coordinate position information of the airborne navigation system in the fixed coordinate system is obtained.
[0108] Step 706, solving the airborne fused line data according to the velocity solving algorithm to obtain the velocity component information of the airborne navigation system in the fixed coordinate system.
[0109] Specifically, the airborne fused line data is input into the velocity solution algorithm, and the velocity component is obtained by calculating the time derivative of the position information through the velocity solution algorithm. This process includes data filtering, error correction and integral calculation, and finally the velocity component information of the airborne navigation system in the fixed coordinate system is obtained.
[0110] Step 708, solving the airborne fused row data according to the heading solution algorithm to obtain the heading component information of the airborne navigation system in the fixed coordinate system.
[0111] Specifically, the airborne fused data is input into the heading solution algorithm, and the heading angle is calculated by fusing the geomagnetic data and angular velocity data through the heading solution algorithm. This process includes data filtering, error correction and angle integral calculation, and finally the heading component information of the airborne navigation system in the fixed coordinate system is obtained.
[0112] Step 710, obtaining airborne navigation parameters according to the coordinate position information, velocity component information and heading component information.
[0113] Specifically, the coordinate position information is used to determine the current position of the navigation system, the velocity component information is used to calculate the motion state of the system, and the heading component information is used to determine the motion direction of the system. By comprehensively processing this information and applying relevant navigation algorithms (such as dead reckoning algorithms), the onboard navigation parameters are calculated and updated, and finally a complete set of onboard navigation parameters is obtained, including key data such as current position, speed and heading.
[0114] In this embodiment, by determining and applying appropriate position solution, velocity solution and heading solution algorithms, accurate positioning, velocity and heading information of the airborne navigation system can be achieved. This method can ensure that the navigation system provides accurate position, velocity and heading parameters in a fixed coordinate system, thereby improving the stability, reliability and accuracy of the navigation system. These parameters are very critical for various applications, such as unmanned driving, aviation navigation, ship navigation, etc., and can help the system achieve accurate path planning, precise position control and safe navigation operations.
[0115] In an exemplary embodiment, Figure 8 As shown, the airborne fused row data is solved according to the position solution algorithm to obtain the coordinate position information of the airborne navigation system in the fixed coordinate system, including:
[0116] Step 802: extract latitude, longitude and altitude information from the airborne fused line data.
[0117] Specifically, the fused navigation data is preprocessed to identify the data fields containing location information. Then, the geographic coordinate data in these fields is parsed to extract the longitude and latitude information. At the same time, the altitude information data field is identified and extracted to ensure that its unit and format are consistent. Through data verification and conversion, accurate latitude, longitude and altitude information are finally obtained.
[0118] Step 804, select any fixed point as the origin of the fixed coordinate system, convert the extracted longitude and latitude information and altitude information into the fixed coordinate system, and obtain initial fixed coordinate information.
[0119] Specifically, any fixed point is selected as the origin of the fixed coordinate system, the longitude, latitude and altitude information of the origin is determined, and the geodetic coordinate conversion algorithm is applied to convert the longitude, latitude and altitude information into X, Y, and Z coordinates in the three-dimensional rectangular coordinate system. The coordinates of the base point are used as a reference to calculate the offset of each point relative to the origin, and finally the initial fixed coordinate information is obtained to achieve accurate positioning by converting the geographic location information to the fixed coordinate system.
[0120] Step 806, calibrating the initial fixed coordinate information according to the ground auxiliary data and the coordinate calibration model corresponding to the airborne navigation system to obtain coordinate position information.
[0121] Among them, the coordinate calibration model can be a mathematical method for accurately aligning and transforming the position and orientation relationship between different coordinate systems. It uses known reference points or reference markers to calculate the transformation parameters (such as translation, rotation, and scaling) between coordinate systems, thereby achieving accurate calibration and alignment of coordinate data.
[0122] Specifically, according to the coordinate calibration model corresponding to the ground auxiliary data and the airborne navigation system, the ground auxiliary data and the initial fixed coordinate information are input into the coordinate calibration model. By comparing the differences between the ground auxiliary data and the initial fixed coordinate information, the initial fixed coordinate information is calibrated using optimization algorithms such as the least squares method. This process includes error calculation, parameter adjustment and coordinate transformation, and finally the calibrated coordinate position information is obtained.
[0123] In this embodiment, by extracting the latitude, longitude and altitude information from the airborne fused navigation data and converting it into initial coordinate information in a fixed coordinate system, the initial processing and conversion of the navigation system position information can be achieved. Selecting a fixed point as the origin of the coordinate system and calibrating the initial coordinate information based on the coordinate calibration model established by ground auxiliary data can improve the accuracy and reliability of the position information. This method can ensure that the position information provided by the navigation system is consistent with the actual geographical location, providing a reliable basis for subsequent applications such as navigation, positioning and map display.
[0124] In an exemplary embodiment, Fig. 9 As shown, the multi-type filter fusion model includes a particle filter algorithm, an unscented Kalman filter algorithm, an integrated learning algorithm, an adaptive information fusion algorithm, and a Bayesian filter algorithm; each airborne filter fusion data includes a particle filter fusion data, an unscented Kalman filter fusion data, an integrated learning fusion data, an adaptive information fusion data, and a Bayesian filter fusion data; Beidou navigation data, inertial navigation data, airborne meteorological data, and ground auxiliary data are input into the multi-type filter fusion model to obtain each airborne filter fusion data, including:
[0125] Step 902, substitute the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the particle filter algorithm, the unscented Kalman filter algorithm, the ensemble learning algorithm, the adaptive information fusion algorithm and the Bayesian filter algorithm to obtain the initial particle filter fusion data, the initial unscented Kalman filter fusion data, the initial ensemble learning fusion data, the initial adaptive information fusion data and the initial Bayesian filter fusion data.
[0126] Specifically, the Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data are respectively input into the particle filter algorithm, the unscented Kalman filter algorithm, the ensemble learning algorithm, the adaptive information fusion algorithm and the Bayesian filter algorithm, and the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data are independently fused by the particle filter algorithm, the unscented Kalman filter algorithm, the ensemble learning algorithm, the adaptive information fusion algorithm and the Bayesian filter algorithm to obtain the initial particle filter fusion data corresponding to the particle filter algorithm, the initial unscented Kalman filter fusion data corresponding to the unscented Kalman filter algorithm, the initial ensemble learning fusion data corresponding to the ensemble learning algorithm, the initial adaptive information fusion data corresponding to the adaptive information fusion algorithm and the initial Bayesian filter fusion data corresponding to the Bayesian filter algorithm.
[0127] Step 904, according to the dynamic game model corresponding to the airborne navigation system, the initial particle filter fusion data, the initial unscented Kalman filter fusion data, the initial integrated learning fusion data, the initial adaptive information fusion data and the initial Bayesian filter fusion data are adjusted to obtain particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data.
[0128] Among them, the dynamic game model can be a mathematical framework for analyzing strategic interactions in multi-stage decision-making processes. It studies the strategic selection and outcome evolution by describing the process of participants making decisions based on each other's behavior and historical information at multiple time points. The model takes into account the decision-making order of the participants, information asymmetry, and time dimension, enabling it to more accurately simulate real-world competition and cooperation scenarios.
[0129] Specifically, the initial particle filter fusion data, the initial unscented Kalman filter fusion data, the initial integrated learning fusion data, the initial adaptive information fusion data and the initial Bayesian filter fusion data are input into the dynamic game model. Through the strategy optimization mechanism in the dynamic game model, each type of data is analyzed for game analysis, and the optimal weights and adjustment parameters between the data are calculated. This includes establishing a game strategy, evaluating the payoff function of each data, and iteratively solving the Nash equilibrium. Finally, the initial data is adjusted to obtain the optimized particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data.
[0130] In this embodiment, by substituting Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into particle filter algorithm, unscented Kalman filter algorithm, integrated learning algorithm, adaptive information fusion algorithm and Bayesian filter algorithm, initial fusion data is obtained, and then the dynamic game model corresponding to the airborne navigation system is used to optimize and adjust these initial fusion data, so as to achieve efficient fusion and accurate calibration of multi-source data. This method not only improves the accuracy and robustness of navigation data, but also can dynamically adapt to environmental changes, and enhance the overall performance and reliability of the navigation system.
[0131] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.
[0132] Based on the same inventive concept, the embodiment of the present application also provides a Beidou airborne data fusion device for implementing the Beidou airborne data fusion method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more Beidou airborne data fusion device embodiments provided below can refer to the above limitations on a Beidou airborne data fusion method, which will not be repeated here.
[0133] In an exemplary embodiment, Fig.10As shown, a Beidou airborne data fusion device is provided, including: a navigation data acquisition module 1002, a fusion model construction module 1004, a fusion data acquisition module 1006, a navigation data acquisition module 1008 and a navigation parameter acquisition module 1010, wherein:
[0134] The navigation data acquisition module 1002 is used to acquire Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system;
[0135] The fusion model building module 1004 is used to build a multi-type filter fusion model corresponding to the airborne navigation system according to Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data;
[0136] The fusion data obtaining module 1006 is used to input Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into a multi-type filter fusion model to obtain each airborne filter fusion data;
[0137] The navigation data obtaining module 1008 is used to fuse the airborne filtered fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system;
[0138] The navigation parameter obtaining module 1010 is used to perform navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system.
[0139] In one embodiment, the navigation data obtaining module 1008 is further used to determine navigation data fusion strategy information according to feature information of each airborne filter fusion data; and fuse each airborne filter fusion data according to the navigation data fusion strategy information to obtain airborne fusion navigation data.
[0140] In one embodiment, the navigation data obtaining module 1008 is also used to determine each navigation data fusion algorithm corresponding to the airborne navigation system when the navigation data fusion strategy information indicates that the fusion mode is a hybrid strategy fusion; according to each navigation data fusion algorithm, each airborne filter fusion data is fused separately to obtain each initial airborne fusion navigation data; when the difference between the fusion navigation parameter curve of any initial airborne fusion navigation data and the current navigation parameter curve is greater than a threshold, the parameters of each navigation data fusion algorithm are adjusted, and the execution is returned to the execution according to each navigation data fusion algorithm, each airborne filter fusion data is fused separately to obtain each initial airborne fusion navigation data, until the difference between the fusion navigation parameter curve and the current navigation parameter curve is less than a threshold, and the airborne fusion navigation data is obtained according to each initial airborne fusion navigation data.
[0141] In one embodiment, the navigation data acquisition module 1008 is also used to select several target filtering algorithms from various fusion filtering algorithms of the multi-type filtering fusion model according to the layer-by-layer fusion algorithm; each target filtering algorithm includes a Gaussian process fusion algorithm, a sparse Bayesian learning algorithm, a deep reinforcement learning fusion algorithm and an integrated Kalman filtering algorithm; according to the fusion step information of the layer-by-layer fusion algorithm, the Gaussian process fusion algorithm is used to fuse the particle filter fusion data and the unscented Kalman filtering fusion data to obtain the first layer of fusion data; according to the fusion step information of the layer-by-layer fusion algorithm, the sparse Bayesian learning algorithm is used to fuse the first layer of fusion data and the integrated learning fusion data to obtain the second layer of fusion data; according to the fusion step information of the layer-by-layer fusion algorithm, the deep reinforcement learning fusion algorithm is used to fuse the second layer of fusion data and the adaptive information fusion data to obtain the third layer of fusion data; according to the fusion step information of the layer-by-layer fusion algorithm, the integrated Kalman filtering algorithm is used to fuse the third layer of fusion data and the Bayesian filtering fusion data to obtain the initial airborne fusion navigation data.
[0142] In one embodiment, the navigation data obtaining module 1008 is also used to fuse any two selected initial airborne fusion navigation data to obtain intermediate fusion airborne navigation data; perform error correction on the intermediate fusion airborne navigation data according to ground auxiliary data to obtain intermediate corrected airborne navigation data; use the intermediate corrected airborne navigation data as one of the selected initial airborne fusion navigation data, select any initial airborne fusion navigation data from the unselected initial airborne fusion navigation data, return to execute the step of fusing any two selected initial airborne fusion navigation data to obtain intermediate fusion airborne navigation data, until all the initial airborne fusion navigation data have been selected at least once to obtain the airborne fusion navigation data.
[0143] In one embodiment, the navigation parameter obtaining module 1010 is also used to determine the position solution algorithm, speed solution algorithm and heading solution algorithm corresponding to the airborne fused line data; the airborne fused line data is solved according to the position solution algorithm to obtain the coordinate position information of the airborne navigation system in the fixed coordinate system; the airborne fused line data is solved according to the speed solution algorithm to obtain the speed component information of the airborne navigation system in the fixed coordinate system; the airborne fused line data is solved according to the heading solution algorithm to obtain the heading component information of the airborne navigation system in the fixed coordinate system; the airborne navigation parameters are obtained according to the coordinate position information, speed component information and heading component information.
[0144] In one embodiment, the navigation parameter obtaining module 1010 is also used to extract longitude and latitude information and altitude information from the airborne fusion line data; select any fixed point as the origin of the fixed coordinate system, convert the extracted longitude and latitude information and altitude information into the fixed coordinate system, and obtain each initial fixed coordinate information; calibrate the initial fixed coordinate information according to the coordinate calibration model corresponding to the ground auxiliary data and the airborne navigation system to obtain the coordinate position information.
[0145] In one embodiment, the fusion data obtaining module 1006 is also used to substitute Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into the particle filter algorithm, unscented Kalman filter algorithm, integrated learning algorithm, adaptive information fusion algorithm and Bayesian filter algorithm to obtain initial particle filter fusion data, initial unscented Kalman filter fusion data, initial integrated learning fusion data, initial adaptive information fusion data and initial Bayesian filter fusion data; according to the dynamic game model corresponding to the airborne navigation system, the initial particle filter fusion data, the initial unscented Kalman filter fusion data, the initial integrated learning fusion data, the initial adaptive information fusion data and the initial Bayesian filter fusion data are adjusted to obtain particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data.
[0146] Each module in the above-mentioned Beidou airborne data fusion device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.
[0147] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.11As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store server data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a Beidou airborne data fusion method is implemented.
[0148] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0149] In one embodiment, a Beidou airborne data fusion system is also provided, the system comprising: a Beidou navigation terminal, an inertial navigation device, a meteorological collection terminal, a ground data transmission terminal and a server;
[0150] Beidou navigation terminal, used to obtain Beidou navigation data and send the Beidou navigation data to the server;
[0151] Inertial navigation device, used to obtain inertial navigation data and send the inertial navigation data to the server;
[0152] Meteorological collection terminal, used to obtain airborne meteorological data and send the airborne meteorological data to the server;
[0153] A ground data sending terminal is used to send the collected ground auxiliary data to the server;
[0154] The server is used to construct a multi-type filter fusion model corresponding to the airborne navigation system based on Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data;
[0155] The server is also used to input Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data into a multi-type filter fusion model to obtain each airborne filter fusion data;
[0156] The server is also used to fuse the airborne filter fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system;
[0157] The server is also used to perform navigation calculation on the airborne fused line data to obtain the airborne navigation parameters corresponding to the airborne navigation system.
[0158] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.
[0159] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.
[0160] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.
[0161] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.
[0162] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.
[0163] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0164] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A Beidou airborne data fusion method, characterized in that: The method comprises: Obtain Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; Constructing a multi-type filtering fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; Inputting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; Fusing the airborne filtered fusion data to obtain airborne fused navigation data corresponding to the airborne navigation system; The fusing of the airborne filtered fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system includes: Determining navigation data fusion strategy information according to characteristic information of each of the airborne filtered fusion data; According to the navigation data fusion strategy information, fusing the airborne filtered fusion data to obtain the airborne fusion navigation data; Wherein, fusing the airborne filtered fusion data according to the navigation data fusion strategy information to obtain the airborne fused navigation data includes: In the case where the navigation data fusion strategy information characterizing the fusion mode is a hybrid strategy fusion, determining each navigation data fusion algorithm corresponding to the airborne navigation system; According to the navigation data fusion algorithms, the airborne filtered fusion data are respectively fused to obtain initial airborne fusion navigation data; In the case where the difference between the fused navigation parameter curve of any of the initial airborne fused navigation data and the current navigation parameter curve is greater than a threshold, adjusting the parameters of each of the navigation data fusion algorithms, and returning to execute the navigation data fusion algorithms to fuse each of the airborne filtered fusion data to obtain each initial airborne fused navigation data, until the difference between the fused navigation parameter curve and the current navigation parameter curve is less than a threshold, and obtaining the airborne fused navigation data according to each of the initial airborne fused navigation data; Wherein, each of the airborne filter fusion data includes particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data; in the case where one of the navigation data fusion algorithms is a layer-by-layer fusion algorithm; according to each of the navigation data fusion algorithms, each of the airborne filter fusion data is fused respectively to obtain each initial airborne fusion navigation data, including: According to the layer-by-layer fusion algorithm, a number of target filtering algorithms are selected from the various fusion filtering algorithms of the multi-type filtering fusion model; each of the target filtering algorithms includes a Gaussian process fusion algorithm, a sparse Bayesian learning algorithm, a deep reinforcement learning fusion algorithm, and an integrated Kalman filtering algorithm; According to the fusion step information of the layer-by-layer fusion algorithm, the particle filter fusion data and the unscented Kalman filter fusion data are fused using the Gaussian process fusion algorithm to obtain the first layer of fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the sparse Bayesian learning algorithm is used to fuse the first layer fusion data and the integrated learning fusion data to obtain the second layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the second layer fusion data and the adaptive information fusion data are fused using the deep reinforcement learning fusion algorithm to obtain the third layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the third layer fusion data and the Bayesian filter fusion data are fused using the integrated Kalman filter algorithm to obtain the initial airborne fusion navigation data; Performing navigation calculation on the airborne fused row data to obtain airborne navigation parameters corresponding to the airborne navigation system; The performing navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system includes: Determine the position solution algorithm, speed solution algorithm and heading solution algorithm corresponding to the airborne fusion line data; Solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused line data according to the velocity solving algorithm to obtain velocity component information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused row data according to the heading solution algorithm to obtain the heading component information of the airborne navigation system in a fixed coordinate system; Obtaining the airborne navigation parameters according to the coordinate position information, the velocity component information and the heading component information; The step of solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system includes: Extracting latitude and longitude information and altitude information from the airborne fused line data; Select any fixed point as the origin of the fixed coordinate system, transform the extracted longitude and latitude information and altitude information into the fixed coordinate system, and obtain each initial fixed coordinate information; The initial fixed coordinate information is calibrated according to the ground auxiliary data and a coordinate calibration model corresponding to the airborne navigation system to obtain the coordinate position information.
2. The method according to claim 1, characterized in that: The obtaining the airborne fused navigation data according to each of the initial airborne fused navigation data comprises: Fusing any two selected initial airborne fused navigation data to obtain intermediate fused airborne navigation data; According to the ground auxiliary data, error correction is performed on the intermediate fused airborne navigation data to obtain intermediate corrected airborne navigation data; The intermediate corrected airborne navigation data is used as one of the selected initial airborne fusion navigation data, any initial airborne fusion navigation data is selected from the unselected initial airborne fusion navigation data, and the step of fusing any two selected initial airborne fusion navigation data to obtain the intermediate fusion airborne navigation data is returned to the step of obtaining the intermediate fusion airborne navigation data, until all the initial airborne fusion navigation data are selected at least once to obtain the airborne fusion navigation data.
3. The method according to claim 1, characterized in that The multi-type filter fusion model includes a particle filter algorithm, an unscented Kalman filter algorithm, an integrated learning algorithm, an adaptive information fusion algorithm and a Bayesian filter algorithm; each of the airborne filter fusion data includes particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data; the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data are input into the multi-type filter fusion model to obtain each airborne filter fusion data, including: Substituting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the particle filter algorithm, the unscented Kalman filter algorithm, the ensemble learning algorithm, the adaptive information fusion algorithm and the Bayesian filter algorithm to obtain initial particle filter fusion data, initial unscented Kalman filter fusion data, initial ensemble learning fusion data, initial adaptive information fusion data and initial Bayesian filter fusion data; According to the dynamic game model corresponding to the airborne navigation system, the initial particle filter fusion data, the initial unscented Kalman filter fusion data, the initial integrated learning fusion data, the initial adaptive information fusion data and the initial Bayesian filter fusion data are adjusted to obtain the particle filter fusion data, the unscented Kalman filter fusion data, the integrated learning fusion data, the adaptive information fusion data and the Bayesian filter fusion data.
4. A Beidou airborne data fusion system, characterized in that: The system includes: a Beidou navigation terminal, an inertial navigation device, a meteorological collection terminal, a ground data transmission terminal and a server; The Beidou navigation terminal is used to obtain Beidou navigation data and send the Beidou navigation data to the server; The inertial navigation device is used to obtain inertial navigation data and send the inertial navigation data to the server; The meteorological collection terminal is used to obtain airborne meteorological data and send the airborne meteorological data to the server; The ground data sending terminal is used to send the collected ground auxiliary data to the server; The server is used to construct a multi-type filtering fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; The server is further used to input the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; The server is further used to fuse the airborne filtering fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system; The fusing of the airborne filtered fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system includes: Determining navigation data fusion strategy information according to characteristic information of each of the airborne filtered fusion data; According to the navigation data fusion strategy information, fusing the airborne filtered fusion data to obtain the airborne fusion navigation data; Wherein, fusing the airborne filtered fusion data according to the navigation data fusion strategy information to obtain the airborne fused navigation data includes: In the case where the navigation data fusion strategy information characterizing the fusion mode is a hybrid strategy fusion, determining each navigation data fusion algorithm corresponding to the airborne navigation system; According to the navigation data fusion algorithms, the airborne filtered fusion data are respectively fused to obtain initial airborne fusion navigation data; In the case where the difference between the fused navigation parameter curve of any of the initial airborne fused navigation data and the current navigation parameter curve is greater than a threshold, adjusting the parameters of each of the navigation data fusion algorithms, and returning to execute the navigation data fusion algorithms to fuse each of the airborne filtered fusion data to obtain each initial airborne fused navigation data, until the difference between the fused navigation parameter curve and the current navigation parameter curve is less than a threshold, and obtaining the airborne fused navigation data according to each of the initial airborne fused navigation data; Wherein, each of the airborne filter fusion data includes particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data; in the case where one of the navigation data fusion algorithms is a layer-by-layer fusion algorithm; according to each of the navigation data fusion algorithms, each of the airborne filter fusion data is fused respectively to obtain each initial airborne fusion navigation data, including: According to the layer-by-layer fusion algorithm, a number of target filtering algorithms are selected from the various fusion filtering algorithms of the multi-type filtering fusion model; each of the target filtering algorithms includes a Gaussian process fusion algorithm, a sparse Bayesian learning algorithm, a deep reinforcement learning fusion algorithm, and an integrated Kalman filtering algorithm; According to the fusion step information of the layer-by-layer fusion algorithm, the particle filter fusion data and the unscented Kalman filter fusion data are fused using the Gaussian process fusion algorithm to obtain the first layer of fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the sparse Bayesian learning algorithm is used to fuse the first layer fusion data and the integrated learning fusion data to obtain the second layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the second layer fusion data and the adaptive information fusion data are fused using the deep reinforcement learning fusion algorithm to obtain the third layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the third layer fusion data and the Bayesian filter fusion data are fused using the integrated Kalman filter algorithm to obtain the initial airborne fusion navigation data; The server is further used to perform navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system; The performing navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system includes: Determine the position solution algorithm, speed solution algorithm and heading solution algorithm corresponding to the airborne fusion line data; Solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused line data according to the velocity solving algorithm to obtain velocity component information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused row data according to the heading solution algorithm to obtain the heading component information of the airborne navigation system in a fixed coordinate system; Obtaining the airborne navigation parameters according to the coordinate position information, the velocity component information and the heading component information; The step of solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system includes: Extracting latitude and longitude information and altitude information from the airborne fused line data; Select any fixed point as the origin of the fixed coordinate system, transform the extracted longitude and latitude information and altitude information into the fixed coordinate system, and obtain each initial fixed coordinate information; The initial fixed coordinate information is calibrated according to the ground auxiliary data and a coordinate calibration model corresponding to the airborne navigation system to obtain the coordinate position information.
5. A Beidou airborne data fusion device, characterized in that: The device comprises: The navigation data acquisition module is used to obtain Beidou navigation data, inertial navigation data, airborne meteorological data and ground auxiliary data corresponding to the airborne navigation system; A fusion model building module, used to build a multi-type filtering fusion model corresponding to the airborne navigation system according to the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data; A fusion data obtaining module, used for inputting the Beidou navigation data, the inertial navigation data, the airborne meteorological data and the ground auxiliary data into the multi-type filter fusion model to obtain each airborne filter fusion data; A navigation data obtaining module, used for fusing the airborne filtered fusion data to obtain airborne fused navigation data corresponding to the airborne navigation system; The fusing of the airborne filtered fusion data to obtain the airborne fusion navigation data corresponding to the airborne navigation system includes: Determining navigation data fusion strategy information according to characteristic information of each of the airborne filtered fusion data; According to the navigation data fusion strategy information, fusing the airborne filtered fusion data to obtain the airborne fusion navigation data; Wherein, fusing the airborne filtered fusion data according to the navigation data fusion strategy information to obtain the airborne fused navigation data includes: In the case where the navigation data fusion strategy information characterizing the fusion mode is a hybrid strategy fusion, determining each navigation data fusion algorithm corresponding to the airborne navigation system; According to the navigation data fusion algorithms, the airborne filtered fusion data are respectively fused to obtain initial airborne fusion navigation data; In the case where the difference between the fused navigation parameter curve of any of the initial airborne fused navigation data and the current navigation parameter curve is greater than a threshold, adjusting the parameters of each of the navigation data fusion algorithms, and returning to execute the navigation data fusion algorithms to fuse each of the airborne filtered fusion data to obtain each initial airborne fused navigation data, until the difference between the fused navigation parameter curve and the current navigation parameter curve is less than a threshold, and obtaining the airborne fused navigation data according to each of the initial airborne fused navigation data; Wherein, each of the airborne filter fusion data includes particle filter fusion data, unscented Kalman filter fusion data, integrated learning fusion data, adaptive information fusion data and Bayesian filter fusion data; in the case where one of the navigation data fusion algorithms is a layer-by-layer fusion algorithm; according to each of the navigation data fusion algorithms, each of the airborne filter fusion data is fused respectively to obtain each initial airborne fusion navigation data, including: According to the layer-by-layer fusion algorithm, a number of target filtering algorithms are selected from the various fusion filtering algorithms of the multi-type filtering fusion model; each of the target filtering algorithms includes a Gaussian process fusion algorithm, a sparse Bayesian learning algorithm, a deep reinforcement learning fusion algorithm, and an integrated Kalman filtering algorithm; According to the fusion step information of the layer-by-layer fusion algorithm, the particle filter fusion data and the unscented Kalman filter fusion data are fused using the Gaussian process fusion algorithm to obtain the first layer of fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the sparse Bayesian learning algorithm is used to fuse the first layer fusion data and the integrated learning fusion data to obtain the second layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the second layer fusion data and the adaptive information fusion data are fused using the deep reinforcement learning fusion algorithm to obtain the third layer fusion data; According to the fusion step information of the layer-by-layer fusion algorithm, the third layer fusion data and the Bayesian filter fusion data are fused using the integrated Kalman filter algorithm to obtain the initial airborne fusion navigation data; A navigation parameter obtaining module, used for performing navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system; The performing navigation calculation on the airborne fused line data to obtain airborne navigation parameters corresponding to the airborne navigation system includes: Determine the position solution algorithm, speed solution algorithm and heading solution algorithm corresponding to the airborne fusion line data; Solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused line data according to the velocity solving algorithm to obtain velocity component information of the airborne navigation system in a fixed coordinate system; Solving the airborne fused row data according to the heading solution algorithm to obtain the heading component information of the airborne navigation system in a fixed coordinate system; Obtaining the airborne navigation parameters according to the coordinate position information, the velocity component information and the heading component information; The step of solving the airborne fused row data according to the position solving algorithm to obtain the coordinate position information of the airborne navigation system in a fixed coordinate system includes: Extracting latitude and longitude information and altitude information from the airborne fused line data; Select any fixed point as the origin of the fixed coordinate system, transform the extracted longitude and latitude information and altitude information into the fixed coordinate system, and obtain each initial fixed coordinate information; The initial fixed coordinate information is calibrated according to the ground auxiliary data and a coordinate calibration model corresponding to the airborne navigation system to obtain the coordinate position information.
Citation Information
Patent Citations
Method for determining carrier acceleration at high precision on basis of multiple ground base stations
CN103389502A
BD / DNS / IMU autonomous integrated navigation system and method thereof
CN103487822A
Integrated navigation method and equipment based on multisource information fusion
CN105758401A
Application method of combination of Beidou navigation and inertial navigation in power grid inspection unmanned aerial vehicle
CN118111426A
Electronic system for dynamic, quasi-realtime measuring and identifying driver maneuvers solely based on mobile phone telemetry, and a corresponding method thereof
WO2019048070A1
