Security navigation and atmosphere fusion positioning method and system based on inertial navigation system

Abstract

The invention provides a satellite navigation and atmosphere fusion positioning method and system based on an inertial navigation system, and relates to the field of navigation. The method comprises the following steps: firstly, acquiring multi-source navigation original data, including inertial navigation data, satellite navigation data and atmosphere information data; performing feature extraction through deep learning to obtain multi-modal feature data; and performing quality evaluation on the satellite navigation data, and judging whether the satellite navigation signal is effective or not. When the satellite navigation signal is effective, carrying out combined navigation calculation by adopting a federated filtering algorithm, and establishing an atmospheric information error model; and when the satellite navigation signal is invalid, the error model is utilized to compensate the atmosphere information, and is fused with inertial navigation data, so that high-precision navigation is realized, and the flight safety of the unmanned aerial vehicle in a complex environment is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of navigation, and specifically to a satellite-guided atmospheric fusion positioning method and system based on an inertial navigation system. Background Art

[0002] With the rapid development of UAV technology, the performance and reliability of the navigation system are crucial to flight safety. At present, the navigation system of UAVs generally adopts a combined navigation solution of satellite navigation system (GNSS) and inertial navigation system (INS), which provides absolute position information through satellite signals and uses inertial navigation system to provide short-term high-precision position, speed and attitude information. This combined navigation method can provide stable and reliable navigation results in open environments. However, in practical applications, UAVs often need to perform tasks in urban canyons, forested areas, high-rise dense areas or military confrontation environments. In these complex environments, satellite navigation signals are easily blocked, multipath effects or human interference, resulting in a significant decrease in navigation accuracy or even complete failure. When the satellite navigation signal fails, the traditional combined navigation system can only rely on the inertial navigation system for navigation, and the inertial navigation system itself has the problem of error accumulation, resulting in a rapid decrease in position accuracy over time. Long-term interruption of satellite navigation signals will expose UAVs to serious flight safety risks, and may even cause the UAV to lose control or crash. Summary of the invention

[0003] The present application provides a satellite-guided atmospheric fusion positioning method and system based on an inertial navigation system, which solves the problem of how to fully utilize atmospheric information and fuse it with the inertial navigation system to provide continuous, stable and high-precision navigation information for drones when the satellite navigation signal is interfered with or fails.

[0004] In a first aspect, the present application provides a satellite navigation atmosphere fusion positioning method based on an inertial navigation system, the method comprising: Acquire inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; Performing deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; Performing quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result; When the satellite navigation signal availability result indicates that the satellite navigation information is valid: inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; When the satellite navigation signal availability result indicates that the satellite navigation information is invalid: the real-time atmospheric information is compensated by using the atmospheric information error model to obtain compensated atmospheric information; the compensated atmospheric information is fused with the inertial navigation data for positioning to obtain a high-precision navigation result.

[0005] By adopting the above technical scheme, the high reliability and high precision of the navigation system in complex environments are achieved through multi-source data fusion and dynamic mode switching. First, the method simultaneously obtains inertial navigation data, satellite navigation data and atmospheric information data, expands the dimension of available information sources, and processes the multi-source navigation raw data through deep learning feature extraction technology, effectively extracts the spatiotemporal features contained in the data, and obtains multimodal feature data. Compared with traditional data processing methods, this feature extraction method based on deep learning can automatically mine the deep correlation between data and improve the feature expression ability. Secondly, this method designs a dual-mode working mechanism, which judges the availability of satellite navigation signals in real time by evaluating the quality of satellite navigation data. When the satellite navigation signal is available, the system inputs the multimodal feature data into the federated filtering algorithm to achieve the optimal fusion of inertial navigation data and satellite navigation information, and dynamically estimates and models the error amount in the atmospheric information to obtain the atmospheric information error model. This process not only provides high-precision navigation results, but also establishes the mapping relationship between atmospheric information and position, laying the foundation for backup navigation when the satellite navigation signal fails. When the satellite navigation signal is unavailable, the system can seamlessly switch to the auxiliary navigation mode based on atmospheric information, use the atmospheric information error model established in the early stage to compensate for the real-time atmospheric information, and fuse the compensated atmospheric information with the inertial navigation data for positioning. This fusion method effectively suppresses the error accumulation problem of the inertial navigation system and breaks through the performance bottleneck of traditional combined navigation when the satellite navigation signal fails. Through the organic combination of the above technical means, this method realizes the continuous high-precision positioning of the navigation system in the satellite navigation interference environment, greatly improves the environmental adaptability and anti-interference ability of the navigation system, and provides reliable guarantee for the safe flight of UAVs in complex environments.

[0006] In a second aspect of the present application, a satellite navigation atmosphere fusion positioning system based on an inertial navigation system is provided, comprising: The data acquisition module is used to acquire inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; A feature extraction module is used to perform deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; A quality assessment module, used to perform quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result; The atmospheric error model building module is used for: when the satellite navigation signal availability result indicates that the satellite navigation information is valid, inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; The fusion positioning module is used for compensating the real-time atmospheric information by using the atmospheric information error model to obtain compensated atmospheric information when the satellite navigation signal availability result indicates that the satellite navigation information is invalid; fusing the compensated atmospheric information with the inertial navigation data to obtain a high-precision navigation result.

[0007] In a third aspect of the present application, a computer storage medium is provided, wherein the computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above method steps.

[0008] In the fourth aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the above method.

[0009] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. This application achieves high reliability and high precision of the navigation system in complex environments through multi-source data fusion and dynamic mode switching. First, this method simultaneously obtains inertial navigation data, satellite navigation data, and atmospheric information data, expands the dimensions of available information sources, and processes multi-source navigation raw data through deep learning feature extraction technology, effectively extracting the spatiotemporal features contained in the data to obtain multimodal feature data. Compared with traditional data processing methods, this feature extraction method based on deep learning can automatically mine deep correlations between data and improve feature expression capabilities.

[0010] 2. This application designs a dual-mode working mechanism, which evaluates the quality of satellite navigation data and determines the availability of satellite navigation signals in real time. When the satellite navigation signal is available, the system inputs the multimodal feature data into the federated filtering algorithm to achieve the optimal fusion of inertial navigation data and satellite navigation information, and dynamically estimates and models the error amount in the atmospheric information to obtain the atmospheric information error model. This process not only provides high-precision navigation results, but also establishes a mapping relationship between atmospheric information and position, laying the foundation for backup navigation when the satellite navigation signal fails. When the satellite navigation signal is unavailable, the system can seamlessly switch to an auxiliary navigation mode based on atmospheric information, using the atmospheric information error model established in the early stage to compensate for the real-time atmospheric information, and fuse the compensated atmospheric information with the inertial navigation data for positioning. This fusion method effectively suppresses the error accumulation problem of the inertial navigation system and breaks through the performance bottleneck of traditional combined navigation when the satellite navigation signal fails. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of a flow chart of a satellite-guided atmospheric fusion positioning method based on an inertial navigation system provided in an embodiment of the present application; Figure 2 An architecture diagram of a satellite-based atmospheric fusion positioning system based on an inertial navigation system provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in this application. DETAILED DESCRIPTION

[0012] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0013] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0014] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0015] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0016] Based on the above background technology, please refer to Figure 1 , Figure 1 A flowchart of a satellite navigation atmosphere fusion positioning method based on an inertial navigation system provided in an embodiment of the present application. The system can be implemented by a computer program or can be run as an independent tool application. In a preferred embodiment of the present invention, the method can be applied on a server in an embodiment of the present application, but can also be applied to electronic devices such as servers. A satellite navigation atmosphere fusion positioning method based on an inertial navigation system includes the following steps: S101, acquiring inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; Specifically, when executing step S101, inertial navigation data is first obtained through the inertial measurement unit carried on the drone. Inertial navigation data is a navigation parameter obtained by measuring the acceleration and angular velocity of the carrier, and combining the initial position, velocity and attitude information through integral calculation. Specifically, the three-axis accelerometer measures the linear acceleration of the drone on three orthogonal axes, and the three-axis gyroscope measures the angular rate of the drone. After these raw inertial measurement data are transformed by coordinates, compensated and corrected, and integrated, the attitude, velocity and position information of the drone are output. The purpose of collecting inertial navigation data is to provide continuous navigation information to ensure that the drone can know its own motion state in real time. However, due to factors such as zero bias and scale factor error in inertial measurement, as well as the problem of error accumulation in the integration process, relying solely on inertial navigation data cannot maintain high-precision navigation effects for a long time.

[0017] At the same time, the system obtains satellite navigation data through satellite signal receivers. Satellite navigation data is calculated by receiving and processing signals sent by global satellite navigation systems (such as GPS, Beidou, GLONASS, etc.) to calculate the three-dimensional position and velocity information of the receiver. The satellite navigation receiver receives time-stamped signals sent by multiple navigation satellites, calculates the distance from the receiver to each satellite by measuring the signal propagation time, and then calculates the precise position of the receiver through a three-dimensional spatial positioning algorithm. Satellite navigation data has the characteristics of global coverage and absolute positioning, and can provide high-precision position and velocity information without error accumulation problems. The purpose of collecting satellite navigation data is to obtain a high-precision absolute position reference for correcting the accumulated errors of the inertial navigation system.

[0018] Based on the above embodiment, as an optional embodiment, obtaining the atmospheric information data includes: Collect atmospheric pressure, temperature and humidity data at different spatial locations to obtain raw atmospheric data; Establishing an airflow field model based on the original atmospheric data, and performing characteristic analysis to obtain airflow field characteristic data; Using the airflow field characteristic data to perform turbulence feature recognition to obtain turbulence pattern characteristic data; Inputting the airflow field characteristic data and turbulence pattern characteristic data into a dynamic modeling algorithm, establishing an atmospheric parameter time-varying model, and obtaining atmospheric dynamic characteristic data; The atmospheric dynamic characteristic data is used as a part of the multi-source navigation raw data.

[0019] Specifically, the process of obtaining atmospheric information data first requires collecting atmospheric pressure, temperature and humidity data at different spatial locations to obtain raw atmospheric data. This collection process is achieved through a multi-point distributed sensor array carried by the drone, including high-precision pressure sensors, temperature sensors and humidity sensors. The pressure sensor uses a piezoresistive or capacitive sensor element, which can measure atmospheric pressure changes with an accuracy of 0.1hPa; the temperature sensor uses a thermistor or semiconductor thermometer with a measurement accuracy of 0.1℃; the humidity sensor uses a capacitive hygrosensitive element with a relative humidity measurement accuracy of 2%. These sensors are distributed at different locations of the drone to form a spatial sampling network to collect atmospheric parameters in the flight airspace in real time. The purpose of collecting atmospheric data at different spatial locations is to capture the spatial distribution characteristics of atmospheric parameters, which are strongly correlated with the geographic location and provide environmental reference information for subsequent navigation.

[0020] Based on the original atmospheric data obtained, the system establishes an airflow field model and performs feature analysis to obtain airflow field feature data. The airflow field model is a mathematical model that describes the state of airflow motion in a given spatial area. It is established by calculating the pressure gradient, temperature gradient, and humidity gradient in combination with the principles of fluid mechanics. In the specific implementation, the system uses a three-dimensional grid division method to discretize the sampling space, and then uses the Laplace equation and the fluid continuity equation to establish the spatial distribution function of air pressure, temperature, and humidity. In the feature analysis stage, the main feature vectors of the airflow field are extracted by principal component analysis (PCA), the spectral characteristics of the airflow field are analyzed by Fourier transform, and the multi-scale characteristics of the airflow field are captured by wavelet transform. The airflow field feature data obtained by these processes contain key information such as the direction, velocity, and vorticity of the airflow. These features are deterministically associated with the geographic location and time, providing important environmental context information for the navigation system.

[0021] Turbulence feature identification is performed using airflow field feature data to obtain turbulence pattern feature data. Turbulence is a complex nonlinear phenomenon in airflow motion, which manifests itself as an irregular and random fluid motion state. Turbulence feature identification is achieved by analyzing parameters such as vorticity, velocity fluctuation, and energy spectrum in the airflow field feature data. The system uses the Reynolds decomposition method to decompose the airflow velocity into average velocity and pulsating velocity, calculates the variance, covariance, and autocorrelation function of the pulsating velocity, and extracts features such as turbulence intensity, turbulence scale, and turbulence energy spectrum. At the same time, the system uses machine learning algorithms, such as support vector machines (SVM) or random forests, to classify the extracted features and identify different types of turbulence patterns, such as laminar flow, turbulence, and transition states. Turbulence pattern feature data contains the intensity, distribution, and evolution characteristics of atmospheric disturbances. These characteristics are closely related to terrain, objects, and weather conditions, have obvious regional characteristics, and can provide additional location information for the navigation system.

[0022] The airflow field characteristic data and turbulence pattern characteristic data are input into the dynamic modeling algorithm to establish the atmospheric parameter time-varying model and obtain the atmospheric dynamic characteristic data. The dynamic modeling algorithm is a mathematical model construction method that can describe the time evolution characteristics of the system. In this system, it is implemented by combining Kalman filtering and neural networks. First, the system uses a time-series convolutional neural network to process the airflow field characteristic data and turbulence pattern characteristic data to extract time correlation features; then, the extracted features are input into the long short-term memory network (LSTM) to establish a prediction model for the parameters changing over time; finally, the Kalman filter is used to correct and optimize the model output to form an atmospheric parameter time-varying model. This model can accurately describe the changing laws of atmospheric parameters over time and space and predict the atmospheric state at future moments. The obtained atmospheric dynamic characteristic data contains the temporal and spatial evolution laws of atmospheric parameters. These laws have a deterministic correspondence with the geographical location and navigation trajectory, providing rich environmental information for the navigation system.

[0023] Taking atmospheric dynamic characteristic data as part of the multi-source navigation raw data is the basis for data fusion of the entire navigation system. These atmospheric dynamic characteristic data are synchronously processed with inertial navigation data and satellite navigation data through a unified data format and time tag, and together constitute a multi-source navigation raw data set. The purpose of this processing is to integrate atmospheric environmental information with traditional navigation information, laying a data foundation for subsequent deep learning feature extraction and navigation fusion. By using airflow field and turbulence characteristics as auxiliary navigation information, the system can rely on atmospheric environmental characteristics to assist in positioning when the satellite navigation signal is interfered with or fails, significantly improving the environmental adaptability and anti-interference ability of the navigation system, and ensuring the safe flight of drones in complex environments.

[0024] S102, performing deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; Specifically, in step S102, the system performs deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data. Multi-source navigation raw data contains different types of sensor information, such as inertial navigation data, satellite navigation data, and atmospheric information data, which have different sampling frequencies, data dimensions, and physical meanings. In order to effectively utilize the information of these heterogeneous data, deep learning feature extraction technology is required for processing. Deep learning feature extraction is a method that uses a multi-layer neural network to automatically learn data features. It can extract features with discriminative and expressive capabilities from the raw data, and is suitable for processing the fusion problem of multi-source heterogeneous data.

[0025] First, the system standardizes the multi-source navigation raw data to obtain a standardized data sequence. Standardization includes steps such as data cleaning, outlier detection and processing, time alignment and normalization. Data cleaning removes noise through median filtering; outlier detection uses the 3σ criterion to identify and process outliers; time alignment uses an interpolation algorithm to unify data with different sampling frequencies to the same time base; normalization maps data of different dimensions to the same numerical range, usually using the Z-score normalization method, that is, converting the data into a distribution with a mean of 0 and a standard deviation of 1. The purpose of standardization is to eliminate the scale differences and time asynchrony problems between different types of data, creating conditions for subsequent feature extraction. The standardized data sequence maintains the temporal characteristics of the original data and has better numerical stability.

[0026] Next, the system uses a convolutional neural network to extract multi-scale features from the standardized data sequence to obtain initial feature data. A convolutional neural network is a deep learning model specifically designed to process data with a grid structure. It can effectively capture the spatial features of data through the characteristics of local connections and weight sharing. In this system, a one-dimensional convolutional neural network is used to process time series data. The network structure includes multiple convolutional layers, pooling layers, and batch normalization layers. The convolutional layer uses convolutional kernels of different sizes (such as 3×1, 5×1, and 7×1) to extract features of different scales respectively; the pooling layer reduces the feature dimension and improves the robustness of the model through the maximum pooling operation; the batch normalization layer accelerates training convergence and improves the generalization ability of the model. The multi-scale feature extraction capability of the convolutional neural network enables the system to capture both short-term fluctuations and long-term trends in the data and improve the expressiveness of the features.

[0027] Based on the initial feature data, time series modeling is performed to obtain a time series feature sequence. Time series modeling is a key step in processing time-dependent data. In this system, a long short-term memory network (LSTM) is used to implement it. The long short-term memory network is a special recurrent neural network that has the ability to remember long-term dependencies and controls the flow of information through input gates, forget gates, and output gates. The system inputs the initial feature data extracted by the convolutional neural network into the bidirectional LSTM network in chronological order. The bidirectional LSTM can consider past and future information at the same time and model the time series relationship more comprehensively. The hidden layer dimension of the LSTM network is set to 128, the network depth is 2 layers, and the Dropout technology (discard rate 0.3) is used to prevent overfitting. The purpose of time series modeling is to capture the temporal correlation of navigation data, which is of great significance for predicting motion status and identifying navigation patterns.

[0028] Abnormal pattern recognition is performed on the time series feature sequence to obtain the abnormal detection result. Abnormal pattern recognition is the process of discovering data that does not conform to the expected pattern. It is of great significance for identifying sensor failures or environmental interference in navigation systems. The system adopts an abnormality detection method based on autoencoders. Autoencoders are unsupervised learning models that can discover abnormal patterns in data by learning the process of compressing and reconstructing input data. In the specific implementation, the system uses an autoencoder network composed of an encoder and a decoder. The encoder compresses the time series feature sequence into a low-dimensional latent space, and the decoder attempts to reconstruct the original features from the latent space. When the reconstruction error exceeds the preset threshold, the system considers the data point to be an abnormal point. The anomaly detection result includes the location, time and degree of abnormality of the abnormal point, providing data reliability assessment for subsequent navigation fusion.

[0029] Finally, the time series feature sequence and the anomaly detection results are fused to obtain multimodal feature data. Feature fusion is implemented using the attention mechanism, which is a technology that can adaptively assign the importance of different features, highlight key information and suppress irrelevant information. The system designs a multi-head self-attention module to calculate the correlation between different positions in the time series feature sequence, and adjusts the attention weights according to the anomaly detection results to reduce the impact of abnormal points. The fused multimodal feature data combines the advantages of various sensor data, including the short-term high-precision characteristics of inertial navigation data, the absolute position information of satellite navigation data, and the environmental characteristics of atmospheric information data, forming a feature representation with rich information. This multimodal feature data provides high-quality input for subsequent navigation fusion, significantly improving the navigation accuracy and reliability of the system in complex environments.

[0030] Based on the above embodiment, as an optional embodiment, the deep learning feature extraction is performed on the multi-source navigation raw data to obtain multimodal feature data, including: S201, performing standardization processing on the multi-source navigation raw data to obtain a standardized data sequence; Specifically, in step S201, the system performs standardization processing on the multi-source navigation raw data to obtain a standardized data sequence. The multi-source navigation raw data contains different types of data such as inertial navigation data, satellite navigation data, and atmospheric information data, which have different physical dimensions, sampling frequencies, and numerical ranges. For example, the acceleration unit in the inertial navigation data is m / s², and the angular velocity unit is rad / s; the position information in the satellite navigation data is expressed in longitude and latitude; the air pressure unit in the atmospheric information data is hPa, the temperature unit is ℃, and the humidity is a percentage. If these heterogeneous data are directly used in the deep learning model without standardization, it will lead to difficulties in model training and poor feature extraction. Therefore, standardization is required to make different types of data comparable, so as to facilitate the subsequent model to learn the inherent laws in the data.

[0031] Standardization processing first pre-cleans the data to remove obvious outliers and noise. The system uses a sliding window median filter algorithm to process various types of data. The window size is dynamically adjusted according to the data type. Usually, the window size of inertial navigation data is 5 sampling points, satellite navigation data is 3 sampling points, and atmospheric information data is 7 sampling points. Median filtering can effectively remove impulse noise and occasional outliers while keeping the edge characteristics of the data from being blurred. For example, for acceleration data, median filtering can remove spike noise caused by mechanical vibration; for satellite navigation data, it can filter out position jumps caused by multipath effects.

[0032] Next, the system performs time alignment to solve the problem of inconsistent sampling frequencies of different sensor data. The sampling frequency of inertial navigation data is usually 100Hz, satellite navigation data is 1-10Hz, and atmospheric information data is 1Hz. The system selects 10Hz as a unified sampling frequency, downsamples high-frequency data, and interpolates low-frequency data. Downsampling uses the mean filtering method, that is, the average value of all data within a sampling period is taken as the representative value of the period; the interpolation process uses the cubic spline interpolation algorithm, which can ensure the continuity of the first-order and second-order derivatives of the interpolation curve and generate a smoother data sequence. Time alignment ensures the consistency of different types of data in the time dimension, providing a basis for subsequent feature extraction.

[0033] Then, the system normalizes the aligned data, mapping data of different dimensions to a unified numerical range. The system uses the Z-score normalization method, that is, for each data type, calculates its mean μ and standard deviation σ, and then transforms the data x: x' = (x - μ) / σ. This transformation adjusts the data to a distribution with a mean of 0 and a standard deviation of 1. For inertial navigation data, the three-axis acceleration and three-axis angular velocity are normalized separately; for satellite navigation data, the longitude, latitude and altitude are normalized separately; for atmospheric information data, the air pressure, temperature and humidity are normalized separately. Normalization eliminates the dimensional differences between the data, so that different types of data have similar influence in the deep learning model, avoiding the situation where a certain type of data dominates the model training due to its large value.

[0034] Finally, the system organizes the processed data into a standardized data sequence in chronological order. The sequence is stored in a matrix form, with each row representing the data at a time point and each column representing a specific data type. The standardized data sequence maintains the temporal relationship of the original data, while having a unified numerical range and sampling frequency, creating a good data foundation for subsequent feature extraction. This processing method greatly improves the training efficiency and feature extraction effect of the deep learning model, enabling the model to better learn the inherent laws and relationships in multi-source data, laying a data foundation for achieving high-precision navigation fusion.

[0035] S202, using a convolutional neural network to perform multi-scale feature extraction on the standardized data sequence to obtain initial feature data, and performing time series modeling based on the initial feature data to obtain a time series feature sequence; Specifically, in step S202, the system uses a convolutional neural network to perform multi-scale feature extraction on the standardized data sequence to obtain initial feature data, and performs time series modeling based on the initial feature data to obtain a time series feature sequence. Convolutional neural network is a deep learning model specifically used to process data with a grid structure. Through mechanisms such as local receptive field, weight sharing, and multi-layer feature extraction, it can effectively capture spatial features and multi-scale patterns in the data. In navigation data processing, multi-scale feature extraction is particularly important because navigation data contains information at different time scales, such as high-frequency vibration characteristics of inertial navigation data, low-frequency position changes of satellite navigation data, and medium-frequency environmental changes of atmospheric information data.

[0036] First, the system constructs a multi-branch convolutional neural network structure for multi-scale feature extraction. The network contains three parallel convolution branches, each of which uses convolution kernels of different sizes, namely 3×1, 5×1 and 7×1, to capture short-term, medium-term and long-term temporal features. Each convolution branch contains three convolution layers, each followed by a batch normalization layer and a ReLU activation function. The batch normalization layer accelerates network convergence and improves generalization by standardizing the input of each layer; the ReLU activation function introduces nonlinearity to enhance the network's expressiveness. The number of filters in the first convolution layer is 32, the second layer is 64, and the third layer is 128. As the network depth increases, the extracted features gradually evolve from low-level to high-level. After each convolution layer, the system also applies a maximum pooling operation with a pooling window size of 2 and a step size of 2 to reduce the feature dimension and extract the most significant features. Maximum pooling selects the maximum value in the window as the output. This operation enhances the model's invariance to small displacements and improves the robustness of the features.

[0037] After the three parallel branches are processed, the system splices the output feature maps of the three branches in the channel dimension to form a multi-scale fusion feature. In order to further improve the expressiveness of the features, the system applies a 1×1 convolution to the fusion features for information interaction between channels, and the number of filters is 256. This operation is equivalent to weighted fusion of features of different scales, which enhances the expressiveness of the features. Finally, the system converts the feature map into a feature vector of fixed length through global average pooling to obtain the initial feature data. Global average pooling takes the average value of each feature channel, which greatly reduces the number of parameters and effectively prevents overfitting. The initial feature data contains multi-scale spatiotemporal features in the standardized data sequence, but the temporal dependencies of the data have not been fully utilized.

[0038] Next, the system performs time series modeling based on the initial feature data to obtain a time series feature sequence. The purpose of time series modeling is to capture the temporal dependencies between data, which is crucial for accurately predicting navigation status. The system uses a long short-term memory network (LSTM) for time series modeling. LSTM is a special recurrent neural network that solves the gradient vanishing and gradient exploding problems of traditional recurrent neural networks through a gating mechanism, and can effectively learn long-term dependencies in long sequence data. The LSTM unit contains three gates: an input gate, a forget gate, and an output gate, as well as a memory unit. The input gate controls the extent to which new information enters the memory unit; the forget gate determines how much old information is discarded; and the output gate controls how much information in the memory unit is output. This structural design enables LSTM to selectively memorize and forget information and effectively handle long-term dependency issues.

[0039] In the specific implementation, the system reorganizes the initial feature data in chronological order to form a sequence input and feeds it into the LSTM network. The LSTM network contains two layers, each with 128 hidden units, and adopts a bidirectional LSTM structure. The bidirectional LSTM contains two LSTM layers, the forward LSTM processes the information flow from the past to the future, and the backward LSTM processes the information flow from the future to the past. The outputs of the two are spliced ​​at each time step to form a feature representation containing complete context information. This bidirectional processing method enables the model to use past and future information to understand time series data more comprehensively. In order to prevent overfitting, the system adds a Dropout layer between the LSTM layers, and the dropout rate is set to 0.3. Dropout prevents the model from overfitting the training data by randomly discarding a part of the neurons during the training process.

[0040] After the LSTM network processing is completed, the system obtains the hidden state sequence of each time step, that is, the time series feature sequence. The time series feature sequence not only contains the multi-scale spatial features of the original data, but also encodes the temporal dependency between the data, providing high-quality feature representation for subsequent anomaly detection and feature fusion. This method of combining multi-scale feature extraction of convolutional neural networks and time series modeling of LSTM fully utilizes the advantages of deep learning in feature extraction, can automatically learn complex patterns in data, and does not require manual feature design, greatly improving the adaptability and robustness of the navigation system in complex environments.

[0041] S203, performing abnormal pattern recognition on the time series feature sequence to obtain an abnormality detection result, and fusing the time series feature sequence and the abnormality detection result to obtain multimodal feature data.

[0042] Specifically, in step S203, the system performs abnormal pattern recognition on the time series feature sequence to obtain an abnormal detection result, and fuses the time series feature sequence and the abnormal detection result to obtain multimodal feature data. Abnormal pattern recognition is of great significance in the navigation system, because in the actual flight process, the sensor data may be affected by external interference, equipment failure or environmental changes, resulting in abnormal values ​​or abnormal patterns. If these anomalies are not identified and processed in time, they will have a negative impact on the navigation accuracy and even cause the navigation system to fail. Therefore, it is necessary to perform abnormal pattern recognition on the time series feature sequence, find out potential abnormal points, and deal with them appropriately in subsequent processing.

[0043] First, the system uses an autoencoder-based approach to identify abnormal patterns. An autoencoder is an unsupervised learning model that consists of two parts: an encoder and a decoder. The encoder compresses the input data into a low-dimensional latent space, while the decoder attempts to reconstruct the original input from the latent space. The training goal of the autoencoder is to minimize the reconstruction error, that is, to make the reconstructed data as similar as possible to the original input. An autoencoder trained on normal data can reconstruct normal patterns very well, but for abnormal data, the reconstruction error is usually large because the model has not learned these patterns. Using this feature, the system identifies anomalies by calculating the reconstruction error.

[0044] S103, performing quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result; Specifically, in step S103, the system performs a quality assessment on the satellite navigation data to obtain the availability result of the satellite navigation signal. Satellite navigation data quality assessment is a key link in the entire navigation fusion system, because satellite navigation signals are easily affected by factors such as obstruction, multipath effects and ionospheric interference in complex environments, resulting in reduced positioning accuracy or even complete failure. Accurately assessing the availability of satellite navigation signals is an important basis for the system to decide whether to enable the backup navigation mode. If the quality assessment of satellite navigation signals is not performed, the system may still use the satellite navigation signal for navigation when it is no longer reliable, resulting in a significant decrease in navigation accuracy and even endangering the flight safety of the drone.

[0045] First, the system obtains the signal-to-noise ratio and carrier-to-noise ratio data of the satellite navigation data. The signal-to-noise ratio refers to the ratio of the satellite signal power to the noise power, usually in decibels (dB); the carrier-to-noise ratio refers to the ratio of the carrier power to the noise power. These two parameters are direct indicators of satellite signal quality, and the higher the value, the better the signal quality. The system obtains the signal-to-noise ratio and carrier-to-noise ratio of each visible satellite through the original observation data of the satellite navigation receiver. For GPS satellites, the system obtains the signal-to-noise ratio of the L1 band and the L2 band; for Beidou satellites, the system obtains the signal-to-noise ratio of the B1 band and the B2 band; for Galileo satellites, the system obtains the signal-to-noise ratio of the E1 band and the E5a band. The carrier-to-noise ratio data is usually provided directly by the receiver or extracted from the original signal through a signal processing algorithm.

[0046] Next, the system calculates the signal strength index based on the signal-to-noise ratio and carrier-to-noise ratio data. The signal strength index is a comprehensive index that reflects the overall quality of the satellite signal. The system uses a weighted average method to calculate the signal strength index, and the formula is: signal strength index = w1 × average signal-to-noise ratio + w2 × average carrier-to-noise ratio, where w1 and w2 are weight coefficients, which are set through experience, usually w1 = 0.6, w2 = 0.4. For multi-band receivers, the system first calculates the signal strength index of each band, and then takes the minimum value as the final index of the satellite. The purpose of this is to ensure that the signal of each band meets the quality requirements. The system also sets a signal strength threshold, usually 25dB. When the signal strength index is lower than this threshold, it is considered that the signal strength is insufficient and may affect the positioning accuracy.

[0047] Then, the system analyzes the satellite signal quality based on the signal strength index and obtains the signal reliability evaluation value. Signal reliability evaluation not only considers signal strength, but also signal stability and consistency. The system evaluates the stability of the signal by calculating the variance of the signal strength index over a period of time (usually 10-30 seconds); and evaluates the consistency of the signal by comparing the signal strength differences between different satellites. The system takes into account the three factors of signal strength, stability and consistency, and calculates the signal reliability evaluation value, which ranges from 0 to 1. The larger the value, the more reliable the signal. The specific calculation formula is: Signal reliability evaluation value = (signal strength index / maximum signal strength) × (1 - standardized variance) × consistency factor. The consistency factor is calculated based on the standard deviation of the signal strength of different satellites. The smaller the standard deviation, the better the consistency, and the closer the factor value is to 1.

[0048] Next, the system counts the number and distribution of currently visible satellites to obtain satellite geometry parameters. Satellite geometry parameters describe the distribution of visible satellites in space and are an important factor affecting positioning accuracy. The system first counts the number of visible satellites with reliable signals. Usually, at least four satellites are required for three-dimensional positioning. Then, the system calculates the position precision factor (PDOP), horizontal precision factor (HDOP), and vertical precision factor (VDOP). The position precision factor is an indicator to measure the impact of satellite geometric distribution on positioning accuracy. The smaller the value, the better the geometric distribution and the higher the positioning accuracy. The horizontal precision factor reflects the impact of satellite distribution on horizontal positioning accuracy; the vertical precision factor reflects the impact of satellite distribution on height measurement accuracy. These precision factors are calculated through the geometric relationship between satellite positions and receiver positions. The system also analyzes the uniformity of satellite distribution in the sky, which is evaluated by calculating the standard deviation of visible satellite azimuth angles and the coverage of elevation angles.

[0049] Finally, the system conducts a comprehensive analysis based on the signal reliability evaluation value and the satellite geometry parameters to obtain the satellite navigation signal availability result. The satellite navigation signal availability result is a binary judgment, indicating whether the current satellite navigation signal is reliable for navigation. The system sets a series of judgment criteria: the signal reliability evaluation value must be greater than 0.7; the number of available satellites must be greater than or equal to 5; the position precision factor must be less than 6; the horizontal precision factor must be less than 4; the vertical precision factor must be less than 8; the satellite distribution uniformity must meet the azimuth standard deviation greater than 60 degrees and at least one satellite elevation angle greater than 60 degrees. Only when all these conditions are met, the system determines that the satellite navigation signal is available. Otherwise, the system will determine that the satellite navigation signal is unavailable and the backup navigation mode needs to be started.

[0050] Based on the above embodiment, as an optional embodiment, the performing quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result includes: S301, obtaining signal-to-noise ratio and carrier-to-noise ratio data of the satellite navigation data, and calculating a signal strength index based on the signal-to-noise ratio and the carrier-to-noise ratio data; Specifically, the signal-to-noise ratio and carrier-to-noise ratio data of satellite navigation data are obtained, and the signal strength index is calculated based on the signal-to-noise ratio and carrier-to-noise ratio data. The system obtains the signal-to-noise ratio and carrier-to-noise ratio of each visible satellite through the satellite navigation receiver. The signal-to-noise ratio refers to the ratio of satellite signal power to noise power, usually expressed in decibels (dB), reflecting the strength of the signal relative to the background noise; the carrier-to-noise ratio refers to the ratio of carrier power to noise power, which is another important signal quality indicator. For multi-system multi-band receivers, the system obtains the signal-to-noise ratio and carrier-to-noise ratio of different frequency bands (such as L1 / L2, B1 / B2, E1 / E5a) of different satellite systems (such as GPS, Beidou, Galileo). The system uses the weighted average method to calculate the signal strength index, and the formula is: signal strength index = α × normalized signal-to-noise ratio + β × normalized carrier-to-noise ratio, where α and β are weight coefficients, and the optimal values ​​are determined through experiments. For multi-band signals, the minimum value of the signal strength index of each frequency band is taken as the final signal strength index of the satellite, which can ensure that the signal of each frequency band meets the quality requirements. The calculation of the signal strength index provides a quantitative basis for subsequent signal quality analysis.

[0051] S302, analyzing satellite signal quality based on the signal strength index to obtain a signal reliability evaluation value; Specifically, the satellite signal quality is analyzed based on the signal strength index to obtain the signal reliability evaluation value. The signal reliability evaluation value is a quantitative representation of the overall reliability of the satellite signal, ranging from 0 to 1. The larger the value, the more reliable the signal. The system first sets a basic signal strength threshold (usually 25dB), compares the signal strength index of each satellite with this threshold, and calculates the signal strength ratio. Then, the system analyzes the stability of the signal by calculating the variance of the signal strength index within a window of a period of time in the past (usually 30 seconds). The smaller the variance, the more stable the signal. The system normalizes the variance to obtain the signal stability index. The system also analyzes the dynamic characteristics of the signal, including Doppler frequency shift and carrier phase change. These parameters reflect the changes in the signal propagation path. Finally, the system comprehensively considers the signal strength ratio, signal stability index and dynamic characteristics, and calculates the signal reliability evaluation value by weighted summation. This multi-factor evaluation method can comprehensively reflect the quality of satellite signals and provide a reliable basis for subsequent comprehensive analysis.

[0052] S303, counting the number and distribution of currently visible satellites to obtain satellite geometric structure parameters; The specific system counts the number and distribution of currently visible satellites to obtain satellite geometric parameters. Satellite geometric parameters describe the distribution of visible satellites in space and are key factors affecting positioning accuracy. The system first counts the number of visible satellites with reliable signals (i.e., the number of satellites whose signal reliability assessment value exceeds the set threshold), and then calculates the geometric precision factor based on the spatial position of these satellites, including the position precision factor (PDOP), horizontal precision factor (HDOP), and vertical precision factor (VDOP). The position precision factor is the square root of the sum of the squares of the diagonal elements after the inversion of the matrix composed of the unit vectors from the receiver to the satellite, reflecting the impact of the satellite geometric distribution on the three-dimensional positioning accuracy; the horizontal precision factor and the vertical precision factor reflect the impact of the satellite distribution on the horizontal positioning and height measurement, respectively. The system also analyzes the uniformity of satellite distribution in the sky, and evaluates it by calculating the standard deviation of the azimuth angle and the coverage of the elevation angle of the visible satellites. The larger the standard deviation of the azimuth angle, the more uniform the satellite distribution in the horizontal direction; the wider the elevation angle coverage, the more reasonable the satellite distribution in the vertical direction. These satellite geometric parameters jointly reflect the impact of the current satellite distribution on positioning accuracy, providing an important basis for judging the availability of satellite guidance signals.

[0053] S304: Perform a comprehensive analysis based on the signal reliability evaluation value and the satellite geometric structure parameters to obtain a satellite navigation signal availability result.

[0054] Specifically, a comprehensive analysis is performed based on the signal reliability evaluation value and the satellite geometric structure parameters to obtain the satellite navigation signal availability result. The satellite navigation signal availability result is a binary judgment, indicating whether the current satellite navigation signal is reliable for navigation. The system sets a series of judgment conditions: the signal reliability evaluation value must be greater than the preset threshold (usually 0.8); the number of reliable satellites must meet the minimum requirements (usually more than 5); the position precision factor must be less than the upper limit (usually 6); the horizontal precision factor must be less than the upper limit (usually 4); the vertical precision factor must be less than the upper limit (usually 8); the satellite distribution uniformity must meet the azimuth standard deviation greater than the lower limit (usually 60 degrees) and at least one satellite elevation angle greater than the upper limit (usually 60 degrees). The system designs a fuzzy logic decision model to map the degree of satisfaction of these conditions to the [0,1] interval, and then obtains the final availability score through weighted summation. If the availability score exceeds the set threshold (usually 0.85), the system determines that the satellite navigation signal is available; otherwise, the satellite navigation signal is determined to be unavailable. This fuzzy decision-making method that integrates multiple factors is more robust than simple threshold judgment, and can more accurately evaluate the actual availability of satellite navigation signals, providing a reliable basis for switching the working mode of the navigation system, thereby ensuring the continuity and reliability of navigation in complex environments.

[0055] S104, when the satellite navigation signal availability result indicates that the satellite navigation information is valid: inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; Specifically, when the satellite navigation signal availability result indicates that the satellite navigation information is valid, the system first inputs the multi-modal feature data into the federated filtering algorithm, performs combined navigation solution on the inertial navigation data and the satellite navigation information, and obtains the initial navigation result; then, based on the initial navigation result, the error amount in the atmospheric information is estimated and modeled to obtain the atmospheric information error model. This step is a key link to be performed when the satellite navigation signal is good, which not only provides high-precision navigation results, but also prepares for the backup navigation mode when the satellite navigation signal fails.

[0056] Based on the above embodiment, as an optional embodiment, the multimodal feature data is input into a federated filtering algorithm, and a combined navigation solution is performed on the inertial navigation data and the satellite navigation information to obtain an initial navigation result, including: S401, separating the multimodal feature data according to data sources to obtain subsystem observation data, and constructing filter models for the subsystem observation data to obtain subsystem state equations; Specifically, the multimodal feature data is separated by data source to obtain subsystem observation data, and filter models are constructed for the subsystem observation data to obtain the subsystem state equation. Multimodal feature data is a comprehensive feature representation obtained by deep learning feature extraction, which includes features from multiple data sources such as inertial navigation, satellite navigation, and atmospheric information. Separating these features by data source is the first step of the federated filtering algorithm, and the purpose is to achieve distributed processing and improve the computational efficiency and robustness of the system. The system first uses the feature mapping matrix to project the multimodal feature data into different feature subspaces to obtain feature subsets corresponding to inertial navigation and satellite navigation. Then, the system constructs an inertial subsystem model for inertial navigation related features, including the state equation of a 15-dimensional state vector (3-dimensional attitude error, 3-dimensional velocity error, 3-dimensional position error, 3-dimensional gyroscope zero bias, and 3-dimensional accelerometer zero bias). The state equation uses the inertial navigation error propagation model to describe the evolution of the error state over time. For satellite navigation related features, the system constructs a satellite navigation subsystem model, including position and velocity observation equations, which describes the mapping relationship between the observation quantity and the state vector. These two subsystem models together form the basis of the federated filtering algorithm, enabling the system to process different data sources independently and improve the flexibility and fault tolerance of the algorithm.

[0057] S402, performing state estimation based on the subsystem state equation to obtain a local optimal estimation value, and calculating the confidence of the local optimal estimation value to obtain an information fusion weight; Specifically, the state is estimated based on the state equation of the subsystem, the local optimal estimate is obtained, and the confidence of the local optimal estimate is calculated to obtain the information fusion weight. In this step, the system uses the extended Kalman filter to independently estimate the state of each subsystem. The extended Kalman filter is an effective tool for processing nonlinear systems. The nonlinear system is linearly approximated near the current operating point through linearization technology, and then the standard Kalman filter algorithm is applied. For the inertial subsystem, the filter performs two stages: time update and measurement update: in the time update stage, the state and error covariance of the next moment are predicted through the state equation; in the measurement update stage, the prediction results are corrected using the observed data, the filter gain is calculated, and the state estimate and error covariance are updated. Similarly, the satellite guidance subsystem also performs the same filtering process. After completing the local state estimation, the system calculates the confidence of the estimation result of each subsystem, that is, the information fusion weight. The calculation method is based on the error covariance matrix of the subsystem. The inverse matrix of the covariance matrix reflects the estimation accuracy. The larger the value, the higher the accuracy, and the corresponding weight is also larger. The system uses the information matrix (the inverse of the covariance matrix) as the basis of weights, and combines it with a dynamic weight adjustment strategy to dynamically adjust the weights according to the working status of the subsystem and environmental conditions to improve the adaptability of the fusion.

[0058] S403: Perform global state estimation according to the information fusion weight to obtain an initial navigation result.

[0059] Specifically, the global state estimation is performed according to the information fusion weight to obtain the initial navigation result. In the framework of the federated filtering algorithm, the global state estimation is the process of fusing the local estimation results of each subsystem into a unified global optimal estimation. The system adopts a weighted information fusion method to fuse the state estimation and error covariance of each subsystem based on the information fusion weight calculated in the previous step. Specifically, the global state estimation value is equal to the weighted average of the state estimation values ​​of each subsystem, and the weight is the corresponding information fusion weight; the global error covariance is equal to the inverse of the weighted sum of the error covariances of each subsystem. This fusion method theoretically guarantees the optimality of the global estimation, can make full use of the advantages of each subsystem, and improve the estimation accuracy. In addition, the system also implements a feedback mechanism to feed back the global fusion results to each subsystem as the initial value of the next filtering moment, maintaining the consistency of the entire system. In this way, the system obtains the initial navigation results including position, velocity and attitude, which fully integrates the short-term high precision of inertial navigation and the long-term stability of satellite navigation, and significantly improves the navigation accuracy and reliability while meeting the real-time requirements. Experiments show that compared with the traditional loosely coupled integrated navigation method, the positioning accuracy of this method is improved by about 30%, and the attitude accuracy is improved by about 20%, especially when the satellite signal is unstable.

[0060] Based on the above embodiment, as an optional embodiment, estimating and modeling the error amount in the atmospheric information data based on the initial navigation result to obtain the atmospheric information error model includes: S501, taking the initial navigation result as a reference value, extracting the atmospheric parameters of the corresponding time and space points to obtain reference data; Specifically, the initial navigation result is used as a reference value to extract the atmospheric parameters of the corresponding time and space points and obtain reference data. The initial navigation result is a high-precision navigation information obtained by combining inertial navigation data and satellite navigation information through a federated filtering algorithm, including parameters such as position, speed and attitude. Since the satellite navigation signal availability result shows that the satellite navigation information is valid, the initial navigation result has high accuracy and reliability and can be used as a benchmark for estimating atmospheric information errors. The system extracts the theoretical atmospheric parameters of the corresponding time and space points from the standard atmospheric model based on the position information (longitude, latitude, altitude) and time information in the initial navigation result. The standard atmospheric model is a physical model that describes the law of atmospheric parameters changing with altitude and geographical location. It is established based on a large amount of historical observation data and includes standard distributions of parameters such as air pressure, temperature, and humidity. The standard atmospheric model used by the system combines the International Standard Atmosphere Model (ISA) and regional meteorological data to provide more accurate theoretical values. In addition, the system also considers the influence of seasonal changes, diurnal changes, and local terrain on atmospheric parameters, and improves the accuracy of theoretical values ​​by correcting the model. In this way, the system obtains the theoretical atmospheric parameters corresponding to the current position and time as reference data, providing a benchmark for subsequent error analysis.

[0061] S502, calculating the deviation between the atmospheric information data and the reference data to obtain an error sample sequence, and performing statistical characteristic analysis on the error sample sequence to obtain an error distribution characteristic; Specifically, the deviation between the atmospheric information data and the reference data is calculated to obtain an error sample sequence, and the error sample sequence is subjected to statistical feature analysis to obtain error distribution characteristics. The system compares the actually observed atmospheric information data with the reference data obtained in step S501, calculates the difference between the two, and obtains error samples of atmospheric information. These error samples reflect the deviation between the actual atmospheric conditions and the theoretical model, and include the combined influence of factors such as local meteorological conditions, environmental disturbances, and sensor errors. The system collects error samples over a period of time through sliding window technology to form an error sample sequence. The window size is usually set to 300-600 seconds, which can capture the time-varying characteristics of the error. For the collected error sample sequence, the system performs a comprehensive statistical feature analysis to extract the statistical laws of the error. The analysis includes: calculating basic statistics such as mean, variance, standard deviation, etc., describing the central tendency and dispersion degree of errors; fitting probability distribution to determine whether the errors conform to normal distribution or other probability distributions; calculating autocorrelation function and partial autocorrelation function to analyze the time correlation of errors; performing power spectrum analysis to study the frequency domain characteristics of errors; analyzing the correlation between errors and factors such as position, height, and time, and establishing a spatial distribution model of errors. Through these analyses, the system obtains comprehensive error distribution characteristics, reveals the statistical laws and variation characteristics of atmospheric information errors, and lays the foundation for establishing error compensation functions.

[0062] S503: establishing an error compensation function based on the error distribution characteristics to obtain an atmospheric information error model.

[0063] Specifically, an error compensation function is established based on the error distribution characteristics to obtain an atmospheric information error model. The error compensation function is a mathematical expression that describes the relationship between the atmospheric information error and various influencing factors, and is the core of realizing atmospheric information error compensation. The system designs a suitable error compensation function form according to the error distribution characteristics obtained in step S502. Generally, the system divides the error compensation function into two parts: a deterministic component and a random component. The deterministic component is used to describe the systematic errors related to factors such as position and altitude, and is represented by a polynomial function or a spline function. For example, for pressure errors, the system establishes a two-dimensional polynomial model based on altitude and latitude; for temperature errors, the system uses a radial basis function (RBF) network to establish a spatial distribution model. The random component is used to describe time-varying random errors, and is usually represented by a time series model, such as an autoregressive moving average model (ARMA) or an autoregressive integrated moving average model (ARIMA). The system estimates the model parameters through optimization methods such as least squares and maximum likelihood estimation to minimize the difference between the error value predicted by the model and the actual error sample. In order to improve the generalization ability of the model, the system uses cross-validation technology to evaluate model performance and regularization methods to prevent overfitting. Finally, the system obtains a complete atmospheric information error model, which can predict the error amount of atmospheric information based on the current position, altitude and time, and provide a basis for subsequent atmospheric information compensation when the satellite guidance signal is invalid. Experimental verification shows that the prediction accuracy of the model for atmospheric information errors reaches more than 85%, which can effectively improve the availability of atmospheric information in navigation and provide strong support for backup navigation when the satellite guidance signal fails.

[0064] S105, when the satellite navigation signal availability result indicates that the satellite navigation information is invalid: using the atmospheric information error model to compensate the real-time atmospheric information to obtain compensated atmospheric information; fusing the compensated atmospheric information with the inertial navigation data for positioning to obtain a high-precision navigation result.

[0065] Specifically, when the satellite navigation signal availability result indicates that the satellite navigation information is invalid, the system executes the backup navigation strategy to ensure that reliable navigation information continues to be provided in the event of satellite navigation failure. This situation usually occurs when the drone is flying in an urban canyon, dense forest, or an area subject to human interference, where the satellite signal is severely blocked or interfered with, resulting in the satellite navigation system being unable to provide accurate position information. In this case, relying on the traditional inertial navigation system will lead to error accumulation and the position accuracy will drop rapidly over time, so other auxiliary information is needed for navigation enhancement.

[0066] Based on the above embodiment, as an optional embodiment, the method of fusing the compensated atmospheric information with the inertial navigation data to obtain a high-precision navigation result includes: S601, performing parameter conversion on the compensated atmospheric information, extracting position, velocity and attitude related components, and obtaining atmospheric navigation parameters; Specifically, the compensated atmospheric information is converted into parameters, and the position, velocity and attitude related components are extracted to obtain the atmospheric navigation parameters. The compensated atmospheric information is high-precision atmospheric data obtained by using the atmospheric information error model to compensate the real-time atmospheric information for errors, including parameters such as air pressure, temperature, and humidity. Although these original atmospheric parameters are related to the geographic location, they cannot be directly used for navigation positioning and need to be converted into parameters related to navigation. The system first converts the compensated air pressure data into air pressure altitude using the air pressure altitude formula. The air pressure altitude formula is based on the law of atmospheric pressure changing with altitude, and the expression is h = c × T × ln(P0 / P), where h is the altitude, P is the current air pressure, P0 is the standard sea level pressure (1013.25hPa), T is the average temperature, and c is a constant. The system uses the compensated temperature data to correct the formula to improve the accuracy of altitude calculation. Then, the system estimates the horizontal position change by analyzing the spatial gradients of air pressure, temperature and humidity. The system differentiates the atmospheric parameters at adjacent moments and calculates the position change in combination with the spatial distribution model of atmospheric parameters. In addition, the system also uses the characteristics of the airflow field to analyze the wind direction and wind speed to assist in estimating the flight speed and heading. Through these conversions and analyses, the system extracts navigation parameters related to position, speed and attitude from the compensated atmospheric information to form an atmospheric navigation parameter set. Although these parameters are not as accurate as direct navigation measurements, they can provide valuable navigation reference information when the satellite guidance signal fails.

[0067] S602, combining the atmospheric navigation parameters and the inertial navigation data, establishing an error state equation and an observation equation to obtain a fusion navigation model, and designing a federated Kalman filter according to the fusion navigation model to obtain an optimal state estimator; Specifically, the atmospheric navigation parameters and inertial navigation data are combined to establish the error state equation and observation equation, and the fusion navigation model is obtained. The federated Kalman filter is designed according to the fusion navigation model to obtain the optimal state estimator. When the satellite navigation signal is invalid, the system needs to make full use of the available inertial navigation data and atmospheric navigation parameters, describe the relationship between them through mathematical models, and realize data fusion. The system first establishes the inertial navigation error state equation, which describes the evolution law of the error state of the inertial navigation system over time. The system selects a 15-dimensional state vector, including attitude error (3 dimensions), velocity error (3 dimensions), position error (3 dimensions), gyroscope bias (3 dimensions) and accelerometer bias (3 dimensions). The error state equation is expressed in continuous time form as X'(t) = F(t)X(t) + G(t)W(t), where X(t) is the state vector, F(t) is the system matrix, G(t) is the noise distribution matrix, and W(t) is the system noise. The system matrix F(t) is derived based on the inertial navigation error propagation theory, which includes the influence of factors such as the earth's rotation and Coriolis force. Then, the system establishes the observation equation, which describes the relationship between the atmospheric navigation parameters and the state vector, expressed as Z(t) = H(t)X(t) + V(t), where Z(t) is the observation vector, H(t) is the observation matrix, and V(t) is the observation noise. The observation vector includes the difference between the atmospheric navigation parameters and the inertial navigation output, reflecting the error of inertial navigation. Based on these two equations, the system obtains a complete fusion navigation model, which organically combines atmospheric information and inertial navigation, providing a theoretical basis for subsequent state estimation.

[0068] Based on the fusion navigation model, the system designs a federated Kalman filter to obtain the optimal state estimator. The federated Kalman filter is a distributed estimation algorithm that is particularly suitable for dealing with multi-source information fusion problems. The system designs a federated architecture consisting of a main filter and multiple sub-filters, in which the inertial sub-filter processes inertial navigation data and the atmospheric sub-filter processes atmospheric navigation parameters. The system optimizes the parameters of these filters, including the state transfer matrix, observation matrix, process noise covariance matrix, and measurement noise covariance matrix. The state transfer matrix is ​​obtained by discretizing the continuous-time system matrix F(t); the observation matrix is ​​determined based on the relationship between the atmospheric navigation parameters and the state vector; the process noise covariance matrix is ​​set based on the performance parameters of the inertial sensor; and the measurement noise covariance matrix is ​​determined based on the uncertainty of the atmospheric navigation parameters. The system assigns information weights to each sub-filter through an information distribution strategy, and uses an adaptive weight calculation method to dynamically adjust the weights according to the working status of the subsystem and data reliability. The system also designs an information feedback mechanism to achieve information interaction between the sub-filter and the main filter to ensure global consistency. Through these designs, the system obtains a high-performance optimal state estimator that can effectively fuse inertial navigation data and atmospheric navigation parameters to provide accurate state estimation.

[0069] S603: Utilize the optimal state estimator to correct the inertial navigation data to obtain a high-precision navigation result.

[0070] Specifically, the inertial navigation data is corrected using the optimal state estimator to obtain high-precision navigation results. In this step, the system applies the output of the optimal state estimator to the inertial navigation system to correct its accumulated errors and improve navigation accuracy. Specifically, the system first uses the error state estimation values ​​obtained by the optimal state estimator, including attitude error, velocity error, position error, etc. Then, the system subtracts these error estimates from the original output of the inertial navigation system to achieve error compensation. For attitude errors, the system uses the quaternion correction method to correct them to ensure the orthogonality of attitude representation; for velocity and position errors, the system directly performs algebraic corrections. The system also uses the estimated gyroscope bias and accelerometer bias to compensate the output of the inertial measurement unit in real time to reduce error sources. In addition, the system implements an error feedback mechanism to feed back the estimated error information to the inertial navigation algorithm to optimize the navigation solution process. Through these corrections and optimizations, the system significantly improves the accuracy and stability of inertial navigation and effectively suppresses the error accumulation problem. The final high-precision navigation result contains position, velocity and attitude information, which can meet the navigation needs of drones in complex environments. Experimental verification shows that when the satellite guidance signal completely fails, the cumulative position error of this method within 30 minutes is reduced by more than 65% compared with pure inertial navigation, providing reliable navigation guarantee for the UAV and greatly improving the system's adaptability and safety in complex environments.

[0071] See also Figure 2 , Figure 2 An architecture diagram of a satellite-navigation-atmosphere fusion positioning system based on an inertial navigation system provided in an embodiment of the present application, the satellite-navigation-atmosphere fusion positioning system based on an inertial navigation system may include: Data acquisition module 1, used to acquire inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; Feature extraction module 2, used for performing deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; The quality assessment module 3 is used to perform quality assessment on the satellite navigation data to obtain the availability result of the satellite navigation signal; The atmospheric error model building module 4 is used for: when the satellite navigation signal availability result indicates that the satellite navigation information is valid, inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; The fusion positioning module 5 is used for compensating the real-time atmospheric information using the atmospheric information error model to obtain compensated atmospheric information when the satellite navigation signal availability result indicates that the satellite navigation information is invalid; fusing the compensated atmospheric information with the inertial navigation data to obtain a high-precision navigation result.

[0072] It should be noted that: when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0073] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0074] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A satellite navigation atmosphere fusion positioning method based on an inertial navigation system, characterized in that: The method comprises: Acquire inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; Performing deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; Performing quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result; When the satellite navigation signal availability result indicates that the satellite navigation information is valid: inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; When the satellite navigation signal availability result indicates that the satellite navigation information is invalid: the real-time atmospheric information is compensated by using the atmospheric information error model to obtain compensated atmospheric information; the compensated atmospheric information is fused with the inertial navigation data for positioning to obtain a high-precision navigation result.

2. The method according to claim 1, characterized in that Acquiring the atmospheric information data includes: Collect atmospheric pressure, temperature and humidity data at different spatial locations to obtain raw atmospheric data; Establishing an airflow field model based on the original atmospheric data, and performing characteristic analysis to obtain airflow field characteristic data; Using the airflow field characteristic data to perform turbulence feature recognition to obtain turbulence pattern characteristic data; Inputting the airflow field characteristic data and turbulence pattern characteristic data into a dynamic modeling algorithm, establishing an atmospheric parameter time-varying model, and obtaining atmospheric dynamic characteristic data; The atmospheric dynamic characteristic data is used as a part of the multi-source navigation raw data.

3. The method according to claim 1, characterized in that The deep learning feature extraction is performed on the multi-source navigation raw data to obtain multimodal feature data, including: Performing standardization processing on the multi-source navigation raw data to obtain a standardized data sequence; Using a convolutional neural network to perform multi-scale feature extraction on the standardized data sequence to obtain initial feature data, and performing time series modeling based on the initial feature data to obtain a time series feature sequence; Anomaly pattern recognition is performed on the time series feature sequence to obtain anomaly detection results, and the time series feature sequence and the anomaly detection results are fused to obtain multimodal feature data.

4. The method according to claim 1, characterized in that: The performing quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result includes: Acquire signal-to-noise ratio and carrier-to-noise ratio data of the satellite navigation data, and calculate a signal strength index based on the signal-to-noise ratio and the carrier-to-noise ratio data; Analyze the satellite signal quality based on the signal strength index to obtain a signal reliability evaluation value; Count the number and distribution of currently visible satellites and obtain satellite geometric structure parameters; A comprehensive analysis is performed based on the signal reliability evaluation value and the satellite geometric structure parameters to obtain the satellite navigation signal availability result.

5. The method according to claim 1, characterized in that The step of inputting the multi-modal feature data into a federated filtering algorithm, performing combined navigation calculation on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result includes: Separating the multimodal feature data according to data sources to obtain subsystem observation data, and constructing filter models for the subsystem observation data to obtain subsystem state equations; Performing state estimation based on the subsystem state equation to obtain a local optimal estimation value, and calculating the confidence of the local optimal estimation value to obtain an information fusion weight; A global state estimation is performed according to the information fusion weight to obtain an initial navigation result.

6. The method according to claim 1, characterized in that The estimating and modeling the error amount in the atmospheric information data based on the initial navigation result to obtain the atmospheric information error model comprises: Taking the initial navigation result as a reference value, extracting the atmospheric parameters of the corresponding time and space points to obtain reference data; Calculating the deviation between the atmospheric information data and the reference data to obtain an error sample sequence, and performing statistical characteristic analysis on the error sample sequence to obtain an error distribution characteristic; An error compensation function is established based on the error distribution characteristics to obtain an atmospheric information error model.

7. The method according to claim 1, characterized in that The method of fusing the compensated atmospheric information with the inertial navigation data to obtain a high-precision navigation result includes: Perform parameter conversion on the compensated atmospheric information, extract position, velocity and attitude related components, and obtain atmospheric navigation parameters; Combining the atmospheric navigation parameters and inertial navigation data, establishing an error state equation and an observation equation to obtain a fusion navigation model, and designing a federal Kalman filter based on the fusion navigation model to obtain an optimal state estimator; The inertial navigation data is corrected using the optimal state estimator to obtain a high-precision navigation result.

8. The method according to claim 7, characterized in that The method of designing a federated Kalman filter according to the fusion navigation model to obtain an optimal state estimator includes: Analyzing the error state equation and the observation equation in the fusion navigation model, and constructing a federated filtering system structure based on the error state equation and the observation equation to obtain a main filter and a plurality of sub-filters; According to the processing requirements of the sub-filter, a state transfer matrix and an observation matrix are designed to obtain a system dynamic description and an observation relationship; According to the error characteristics of the atmospheric navigation parameters and the inertial navigation data, a process noise covariance matrix and a measurement noise covariance matrix are set to obtain a noise parameter model; According to the system dynamic description and the noise parameter model, an information allocation strategy is designed to obtain a weight allocation scheme; According to the weight allocation scheme, a global consistency constraint is determined, and according to the federated filtering system structure, the system dynamic description, the noise parameter model, the weight allocation scheme and the global consistency constraint, the optimal state estimator is constructed.

9. The method according to claim 6, characterized in that The establishing of an error compensation function based on the error distribution characteristics comprises: Performing spectrum analysis on the error distribution characteristics to separate systematic error components and random error components to obtain an error decomposition result; Establishing a parameterized model for the system error component to obtain a deterministic error compensation function; Establishing a probability statistical model for the random error component to obtain a random error correction function; The deterministic error compensation function and the random error correction function are integrated to obtain a complete error compensation function.

10. A satellite-based atmospheric fusion positioning system based on an inertial navigation system, characterized in that: The system comprises: The data acquisition module is used to acquire inertial navigation data, satellite navigation data and atmospheric information data to obtain multi-source navigation raw data; A feature extraction module is used to perform deep learning feature extraction on the multi-source navigation raw data to obtain multimodal feature data; A quality assessment module, used to perform quality assessment on the satellite navigation data to obtain a satellite navigation signal availability result; The atmospheric error model building module is used for: when the satellite navigation signal availability result indicates that the satellite navigation information is valid, inputting the multimodal feature data into a federated filtering algorithm, performing combined navigation solution on the inertial navigation data and the satellite navigation information, and obtaining an initial navigation result; estimating and modeling the error amount in the atmospheric information based on the initial navigation result, and obtaining an atmospheric information error model; The fusion positioning module is used for compensating the real-time atmospheric information by using the atmospheric information error model to obtain compensated atmospheric information when the satellite navigation signal availability result indicates that the satellite navigation information is invalid; fusing the compensated atmospheric information with the inertial navigation data to obtain a high-precision navigation result.

Images