Intelligent calibration method and system for vehicle-mounted satellite positioning device

Abstract

The present invention relates to the field of intelligent traffic management technology, and in particular to an intelligent verification method and system for a vehicle-mounted satellite positioning device, the method comprising the following steps: constructing a dynamic signal fitting model for the vehicle-mounted satellite positioning device; performing collaborative verification on the vehicle-mounted satellite positioning device based on the dynamic signal fitting model to obtain a preliminary verification result; constructing a verification chain and a storage chain to obtain a credible verification result of the preliminary verification result; constructing a cloud-based digital twin model based on the credible verification result to obtain a digital twin of the vehicle-mounted satellite positioning device; establishing a detection and verification platform for the vehicle-mounted satellite positioning device, detecting the digital twin to obtain an intelligent verification result, and realizing intelligent verification of the vehicle-mounted satellite positioning device. The present invention constructs a closed loop for the entire intelligent verification process of the vehicle-mounted satellite positioning device, significantly improves the verification efficiency and device reliability, and optimizes the intelligent level of road transport safety supervision.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic management, and in particular to an intelligent calibration method and system for a vehicle-mounted satellite positioning device. Background Art

[0002] With the rapid development of intelligent transportation and Internet of Vehicles (IoV) technologies, on-board satellite positioning devices have become core components for vehicle safety monitoring, autonomous driving, and logistics scheduling. However, existing detection technologies still face significant bottlenecks: First, traditional detection methods are mostly limited to module-level single-point testing and lack the ability to simulate dynamic scenarios at the vehicle level. This is especially true in complex environments such as tunnels and urban canyons, where signal attenuation and multipath effects lead to significant positioning errors, and detection results deviate from actual operating conditions. Second, the detection process is highly dependent on manual operation and static standard signal sources, making it difficult to adapt to the requirements of multi-band signal synchronization verification. Furthermore, the heterogeneous nature of operator platform interfaces leads to inefficient data collection. Third, the security and credibility of detection data are insufficient, lacking full-link encryption and traceability mechanisms, making it susceptible to tampering or loss, making it difficult to meet detection and verification requirements.

[0003] In the current technology landscape, vehicle-mounted terminal testing mostly uses a serial single-device testing mode, which prevents efficient parallel testing of multiple devices. Differences in operator platform interface protocols require multiple manual adaptations of test data, which is time-consuming and error-prone. The hybrid storage architecture for hot and cold data restricts the performance of large-scale historical trajectory backtracking and real-time analysis. Furthermore, tracing detection errors relies on empirical judgment and lacks intelligent diagnostic methods, making it impossible to accurately identify hidden faults such as chip aging and antenna offset. Although some solutions incorporate analog signal sources, their fixed scenarios and lagging parameter adjustments make it difficult to dynamically simulate real-world road conditions, resulting in low detection coverage and severely restricting the large-scale application and reliability verification of vehicle-mounted positioning devices.

[0004] In view of this, the present invention achieves verification and detection accuracy in complex environments by constructing a dynamic signal fitting model; combines the verification chain and the evidence chain, and constructs a cloud-based digital twin model and detection and verification platform for the vehicle-mounted satellite positioning device, breaking through the problem of data heterogeneity and ensuring that the detection process is efficient, transparent and complies with functional safety standards; it fills the gaps in existing technologies in parallel detection, adaptive scenario simulation and data trust management, and provides technical support for high-precision positioning and full life cycle management of intelligent connected vehicles. Summary of the Invention

[0005] In view of the defects in the prior art, the present invention provides an intelligent calibration method and system for a vehicle-mounted satellite positioning device.

[0006] In order to achieve the above-mentioned objectives, in a first aspect, the present invention provides an intelligent verification method for a vehicle-mounted satellite positioning device, the method comprising the following steps: constructing a dynamic signal fitting model of the vehicle-mounted satellite positioning device by combining multimodal analog signals and adaptive scene signals; performing collaborative verification on the vehicle-mounted satellite positioning device based on the dynamic signal fitting model to obtain a preliminary verification result; constructing a verification chain and a storage chain, and combining the verification chain and the storage chain to obtain a trusted verification result of the preliminary verification result; constructing a cloud-based digital twin model based on the trusted verification result, and obtaining a digital twin of the vehicle-mounted satellite positioning device according to the cloud-based digital twin model; establishing a detection and verification platform for the vehicle-mounted satellite positioning device, and performing detection on the digital twin based on the detection and verification platform to obtain an intelligent verification result, thereby realizing intelligent verification of the vehicle-mounted satellite positioning device. The present invention improves positioning accuracy and environmental adaptability through multimodal signal fusion and dynamic modeling, and uses a collaborative verification mechanism to enhance the comprehensiveness and accuracy of data verification. It combines the verification chain and the evidence chain to ensure that the verification results are tamper-proof and traceable, and builds a highly reliable data foundation. The digital twin realizes real-time mapping of the device's operating status and reduces the cost of physical testing. The detection and verification platform automatically completes the verification, forming a closed-loop process, significantly improving verification efficiency and device reliability, and providing full-life cycle intelligent protection for vehicle-mounted satellite positioning devices.

[0007] Optionally, the dynamic signal fitting model of the vehicle-mounted satellite positioning device is constructed by combining multimodal simulation signals and adaptive scene signals, including: performing virtual scene fitting based on a simulation signal generating device to obtain the multimodal simulation signal; constructing a three-dimensional road model, mapping the physical environment according to the three-dimensional road model to obtain the adaptive scene signal; and dynamically fusing the multimodal simulation signal and the adaptive scene signal to construct the dynamic signal fitting model. The present invention generates multimodal signals through a simulation signal generating device to virtually reproduce complex electromagnetic environments and extreme scenes, breaking through the limitations of physical testing; the three-dimensional road model accurately maps the topological characteristics of the physical environment, combines real-time traffic flow and weather data to generate adaptive scene signals, and enhances the authenticity of environmental perception; and then constructs a dynamic signal fitting model to achieve the complementarity of virtual data and real data, significantly improving the signal anti-interference ability and multi-source data fusion accuracy of the positioning device in dynamic scenes, providing a simulation environment foundation for subsequent verification, and reducing the risk and cost of actual vehicle testing.

[0008] Optionally, the vehicle-mounted satellite positioning device is collaboratively verified based on the dynamic signal fitting model to obtain a preliminary verification result, including: obtaining a dynamic signal through the dynamic signal fitting model, and fusing the dynamic signal with the original data of the vehicle-mounted satellite positioning device to obtain a fused signal; filtering the fused signal to obtain a corrected positioning trajectory of the vehicle-mounted satellite positioning device; obtaining a standard value of the dynamic signal, and combining the standard value to obtain an error value of the corrected positioning trajectory; setting a dynamic threshold according to the dynamic fitting model, and obtaining a multi-dimensional anomaly detection result of the vehicle-mounted satellite positioning device based on the dynamic threshold; and performing collaborative verification based on the error value and the multi-dimensional anomaly detection result to obtain the preliminary verification result. The present invention generates a high-fidelity dynamic signal through a dynamic signal fitting model, which is fused with the original data and filtered to significantly improve the positioning trajectory accuracy, effectively suppressing interference and noise; based on standard value error analysis, quantitative evaluation of positioning performance is achieved, and the dynamic threshold mechanism can adaptively adjust the anomaly detection sensitivity to cover multi-dimensional fault modes such as electromagnetic interference, signal shielding, and clock drift; the collaborative verification framework combines error tracing with anomaly diagnosis to form a closed-loop feedback mechanism, providing an accurate device performance baseline for trusted verification.

[0009] Optionally, the construction of the verification chain and the evidence chain, and combining the verification chain and the evidence chain to obtain a credible verification result of the preliminary verification result, include: obtaining a time-series evidence chain, and constructing the verification chain based on the time-series evidence chain in combination with distributed verification nodes; establishing a multimodal data chain, and constructing the evidence chain based on the multimodal data chain in combination with a privacy protection mechanism; and cross-checking the preliminary verification result in combination with the verification chain and the evidence chain to obtain the credible verification result. The present invention constructs a verification chain and an evidence chain through blockchain technology to achieve full-link credible traceability of the verification process. The verification chain is based on the collaboration of the time-series evidence chain and distributed nodes to ensure that the verification logic is transparent and cannot be tampered with; the evidence chain integrates multimodal data and privacy protection mechanisms to ensure the complete storage of original data and verification evidence. The cross-verification of the two can automatically compare multi-dimensional verification results, identify data tampering and logical loopholes, form an anti-counterfeiting, auditable closed-loop verification system, and significantly improve the credibility of the preliminary verification results.

[0010] Optionally, the cross-verification of the preliminary verification result by combining the verification chain and the evidence chain to obtain the trusted verification result includes: pre-verification of the preliminary verification result based on the verification chain to obtain a first verification result; performing evidence verification on the first verification result according to the evidence chain to obtain a second verification result; and post-verification of the second verification result according to the verification chain to obtain the trusted verification result. The present invention improves the credibility of the preliminary verification result by constructing a double protection mechanism; pre-verification utilizes the distributed consensus mechanism of the verification chain to eliminate logical fallacies; evidence chain verification is based on multimodal data evidence and privacy protection technology to achieve complete anchoring of original data and operation traces; post-verification forms a closed-loop feedback to eliminate the risk of single point failure.

[0011] Optionally, the cloud-based digital twin model is constructed based on the trusted verification result, and the digital twin of the vehicle-mounted satellite positioning device is obtained according to the cloud-based digital twin model, including: standardizing and extracting features from the trusted verification result to obtain key features, and constructing a feature correlation spectrum based on the key features; constructing a cloud-based digital twin model of the vehicle-mounted satellite positioning device according to the feature correlation spectrum; and performing real-time dynamic mapping of the vehicle-mounted satellite positioning device based on the cloud-based digital twin model to obtain the digital twin. The present invention eliminates data heterogeneity through standardization and feature extraction, providing high-precision input for the twin model; the feature correlation spectrum reveals the deep coupling relationship between parameters such as positioning error, signal interference, and hardware status, so that the cloud-based digital twin model has full-factor mapping capabilities; the real-time dynamic mapping mechanism can synchronize the state evolution of the device and the twin, reducing the cost of physical testing.

[0012] Optionally, the cloud-based digital twin model of the vehicle-mounted satellite positioning device is constructed based on the feature association spectrum, including: establishing a model hierarchical architecture of the cloud-based digital twin model, the model hierarchical architecture including a physical layer, a signal layer, and a behavior layer; based on the feature association spectrum, combined with the model hierarchical architecture, hybrid modeling is performed on the vehicle-mounted satellite positioning device to obtain the cloud-based digital twin model. The present invention achieves virtual mapping of the vehicle-mounted positioning device through hierarchical hybrid modeling; the hierarchical architecture achieves modular decoupling, which facilitates algorithm optimization for specific levels; the hybrid modeling takes into account both physical constraints and data-driven characteristics, cloud-based deployment supports elastic expansion, and automatic parameter tuning is achieved in combination with the feature association spectrum, shortening the verification cycle.

[0013] Optionally, the establishment of a detection and verification platform for the vehicle-mounted satellite positioning device includes: establishing a multi-level platform architecture, the multi-level platform architecture including a terminal layer, a storage layer, a service layer, an application layer, and an interface layer; and obtaining the detection and verification platform based on the multi-level platform architecture. The present invention constructs a detection and verification platform architecture for a vehicle-mounted satellite positioning device, achieving modular management and efficient collaboration through a layered design; the terminal layer directly connects to the device to ensure data collection accuracy; the storage layer provides secure and reliable data management; the service layer encapsulates detection algorithms and data analysis capabilities; the application layer realizes human-computer interaction and result visualization; and the interface layer opens up cross-system data interaction. This improves detection efficiency and scalability, and realizes standardized, automated, and intelligent full-process management.

[0014] Optionally, the detection of the digital twin based on the detection and verification platform obtains an intelligent verification result, and realizes the intelligent verification of the vehicle-mounted satellite positioning device, including: obtaining the typical failure mode of the digital twin based on the cloud digital twin model, and obtaining the sequence signal of the typical failure mode; fitting the positioning behavior of the vehicle-mounted satellite positioning device in combination with the multimodal simulation signal and the sequence signal, and obtaining the metrological parameters of the vehicle-mounted satellite positioning device; and performing comparative verification on the metrological parameters through the detection and verification platform to obtain a detection and verification report as the intelligent verification result of the vehicle-mounted satellite positioning device. The present invention realizes high-precision verification through digital twin technology: based on the injection of sequence signals of typical failure modes, typical failure scenarios are identified and the fault reproduction cycle is shortened; the fusion fitting of multimodal signals and fault sequences reduces the simulation error of positioning behavior; the detection and verification platform automatically completes the comparison of metrological parameters to obtain intelligent verification results; and forms a closed-loop intelligent verification system.

[0015] In a second aspect, the present invention provides an intelligent verification system for a vehicle-mounted satellite positioning device. The system implements the intelligent verification method for a vehicle-mounted satellite positioning device provided herein. The system includes an input device, an output device, a processor, and a memory. Its benefits lie in the following: the hardware facilities integrated in the present invention have excellent performance, the input device, output device, processor, and memory are interconnected, and information is smoothly transmitted between the various components. Through the interaction of multiple hardware facilities, an efficient information processing system is constructed. The present invention establishes an efficient information processing architecture and forms a closed-loop verification mechanism, significantly improving the verification efficiency and accuracy of the vehicle-mounted satellite positioning device, reducing the cost of manual intervention, and providing an intelligent solution for quality assurance of vehicle-mounted positioning devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of an intelligent calibration method for a vehicle-mounted satellite positioning device according to an embodiment of the present invention;

[0017] Figure 2Schematic diagram of the platform architecture of the detection and verification platform according to an embodiment of the present invention;

[0018] Figure 3 This is an operational flow chart of the detection and verification platform according to an embodiment of the present invention;

[0019] Figure 4 This is a framework diagram of an intelligent verification system for a vehicle-mounted satellite positioning device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] Specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the present invention. In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, it will be apparent to one of ordinary skill in the art that these specific details are not necessarily required to practice the present invention. In other instances, well-known circuits, software, or methods are not specifically described to avoid obscuring the present invention.

[0021] Throughout this specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Therefore, appearances of the phrases "in one embodiment," "in an embodiment," "an example," or "an example" in various places throughout this specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, or characteristics may be combined in any suitable combinations and / or subcombinations in one or more embodiments or examples. Furthermore, those of ordinary skill in the art will appreciate that the figures provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0022] See Figure 1 One embodiment of the present invention provides an intelligent calibration method for a vehicle-mounted satellite positioning device, the method comprising the following steps:

[0023] S1. Construct a dynamic signal fitting model for the vehicle-mounted satellite positioning device by combining multimodal simulation signals and adaptive scene signals.

[0024] Among them, S1 specifically includes the following steps:

[0025] S11 . Perform virtual scene fitting based on a simulation signal generating device to obtain the multimodal simulation signal.

[0026] In this embodiment, a parameterized virtual scene library is constructed using a multimodal analog signal generator, linking signal time-domain characteristics with spatial geometric constraints. First, a signal generation module is designed based on the radio architecture, and an analog signal generator is established to support multi-frequency signal generation. Each frequency is independently configured with power spectral density, code phase offset, and Doppler shift parameters. Second, a three-dimensional spatial coordinate system is introduced to geometrically model the virtual scene, mapping environmental parameters such as building height, road curvature, and vegetation obstruction into signal attenuation factors and reflection coefficients.

[0027] Specifically, a multipath signal superposition model is constructed to satisfy the following relationship:

[0028]

[0029] in, For the moment The multimodal analog signal, is the total amount of path signal, is the index variable of the path signal, For the The amplitude of the path signal, is the base of natural logarithms, is the imaginary unit, is the carrier angular frequency, is the initial phase, is the signal attenuation factor, is the propagation distance, is the total amount of obstacles, is the index variable of the obstacle, is the reflection coefficient, is a step function, For the The signal of the first path is The shortest distance between obstacles, is the distance threshold.

[0030] Furthermore, the multipath signal superposition model breaks through the limitations of traditional two-dimensional signal simulation and achieves accurate correspondence between the signal power spectrum and the physical environment through spatial geometric constraints.

[0031] S12: Construct a three-dimensional road model, and map the physical environment according to the three-dimensional road model to obtain the adaptive scene signal.

[0032] In this embodiment, the core of the adaptive scene signal is to construct a high-precision three-dimensional road model and achieve dynamic mapping of the physical characteristics of the environment. Based on high-precision maps and real-time traffic flow data, the three-dimensional road model is constructed, and elements such as lane lines, traffic signs, and building outlines are extracted through point cloud semantic segmentation technology to generate a scene topology map with topological relationships. On this basis, using the measured data of the on-board camera and radar, the environment perception algorithm updates the obstacle position, vehicle density and other parameters in the model in real time, and calculates the dynamic occlusion area based on the physics engine. Based on the current model state, the adaptive scene signal generation module uses a ray tracing algorithm to calculate satellite visibility, obtain real-time atmospheric parameters, and dynamically generate an adaptive scene signal that includes error terms such as ionospheric delay and tropospheric refraction.

[0033] Specifically, the laser radar collects point cloud data to construct a 3D road model, which satisfies the following relationship:

[0034]

[0035] in, is the mathematical representation function of the road, is the total amount of point cloud data, is the index variable of the point cloud data, For the The weight coefficient of point cloud data, For the The three-dimensional coordinates of the point cloud data, For the The spatial diffusion parameter of point cloud data.

[0036] In an optional embodiment, a dynamic road feature matrix is ​​constructed in combination with the geometric parameters of the road, and the dynamic road feature matrix is ​​fused with the adaptive scene signal. The dynamic road feature matrix serves as a supplement to the adaptive scene signal to enhance accuracy.

[0037] Specifically, the above adaptive scene signal satisfies the following relationship:

[0038]

[0039] in, For the moment Adaptive scene signal, is the fusion function, is the mathematical representation function of the road, is the dynamic road feature matrix, For time.

[0040] S13. Dynamically fuse the multimodal simulation signal and the adaptive scene signal to construct the dynamic signal fitting model.

[0041] In this embodiment, a dynamic fusion algorithm is used to dynamically fuse multimodal analog signals with adaptive scene signals. A spatiotemporal attention network is constructed, extracting signal timing and spatial distribution features through a three-dimensional convolutional layer. An adaptive fusion gating mechanism is designed to dynamically adjust the weight coefficients of the multimodal signals based on the complexity of the current scene. During the fusion process, signal consistency indicators are continuously monitored, and when the multipath error exceeds a threshold, the model reconstruction process is automatically triggered. By deeply integrating physical signal generation with digital scene driving, a full-lifecycle testing solution is provided for vehicle-mounted positioning devices.

[0042] Specifically, the quality evaluation value of the multimodal simulation signal is obtained, the confidence index value of the adaptive scene signal is obtained, and the dynamic signal fitting model is constructed by combining the quality evaluation value and the confidence index value. The dynamic signal fitting model satisfies the following relationship:

[0043]

[0044] in, is the dynamic signal to be fitted, is the quality assessment value, For the moment The multimodal analog signal, is the confidence index value, For the moment Adaptive scene signal.

[0045] S2. Perform collaborative verification on the vehicle-mounted satellite positioning device according to the dynamic signal fitting model to obtain a preliminary verification result.

[0046] In this embodiment, a dynamic signal is obtained through a dynamic signal fitting model, and the dynamic signal is fused with the original data of the vehicle-mounted satellite positioning device to obtain a fused signal; the fused signal is filtered to obtain a corrected positioning trajectory of the vehicle-mounted satellite positioning device; a standard value of the dynamic signal is obtained, and the error value of the corrected positioning trajectory is obtained in combination with the standard value; a dynamic threshold is set according to the dynamic fitting model, and a multi-dimensional anomaly detection result of the vehicle-mounted satellite positioning device is obtained based on the dynamic threshold; and a preliminary verification result is obtained by performing a collaborative verification based on the error value and the multi-dimensional anomaly detection result.

[0047] First, a dynamic signal is obtained through a dynamic signal fitting model. The dynamic signal includes but is not limited to multipath effects and non-line-of-sight propagation characteristics in typical scenarios such as urban canyons, viaducts, and tunnel entrances and exits. The dynamic signal is spatiotemporally aligned with the original observation data of the vehicle-mounted satellite positioning device through time synchronization, and a dynamic weight fusion algorithm is used to achieve data fusion. During the fusion process, the weight ratio of the simulated signal and the original data is adjusted in real time according to the quality assessment value and confidence index value. The data fusion introduces the robust estimation theory to ensure that the fused signal suppresses abnormal interference while retaining valid information. The final generated fusion signal combines the dynamic characteristics of the real observation with the scene prior knowledge of the simulated signal, providing high-quality input for subsequent processing.

[0048] Secondly, the fused signal undergoes multi-stage filtering: the first stage utilizes an improved Kalman filter, which adaptively adjusts the process noise covariance matrix to achieve real-time compensation for the vehicle dynamic model error. The filter introduces a road curvature constraint term, transforming road geometry parameters into additional constraints on the state equation, effectively suppressing the trajectory divergence problem of traditional Kalman filtering in curved scenarios. The second stage deploys a particle filter, which resamples particle weights using a dynamic signal fitting model. In non-line-of-sight scenarios such as tunnels and urban canyons, the particle set tends to converge toward the predicted area of ​​the simulated signal, improving positioning continuity. The third stage uses a bidirectional smoothing algorithm, combining future observations to reversely correct historical trajectories, ensuring consistency in the time domain. This three-stage filtering architecture optimizes the positioning trajectory error while ensuring real-time performance, resulting in a corrected positioning trajectory.

[0049] Subsequently, a high-precision reference trajectory is constructed using a laser tracker deployed at the calibration site. This serves as the standard value for the dynamic signal and provides a benchmark for error analysis. The position deviation components between the corrected trajectory and the reference trajectory are calculated as error values, including positioning error, velocity error, mileage error, and azimuth error. Spatiotemporal correlation error analysis is introduced to identify periodic error components caused by multipath effects and trend error components caused by changes in satellite geometric distribution by calculating the autocorrelation and cross-correlation functions of the error sequence. The error report includes an integrity risk indicator, which assesses the probability of the positioning result exceeding the allowable error limit based on the protection level algorithm, providing a quantitative basis for collaborative verification.

[0050] Specifically, the above positioning error satisfies the following relationship:

[0051]

[0052] in, for The positioning error of the point, for Point measurement coordinate, Standard coordinate, for Point measurement coordinate, Standard coordinate;

[0053] The above speed error satisfies the following relationship:

[0054]

[0055] in, is the speed error of the vehicle-mounted satellite positioning device, is the average speed measured by the vehicle-mounted satellite positioning device, To simulate standard speed;

[0056] The above mileage error satisfies the following relationship:

[0057]

[0058] in, is the mileage error of the vehicle-mounted satellite positioning device, The measured mileage of the vehicle-mounted satellite positioning device, To simulate standard mileage;

[0059] The above directional angle error satisfies the following relationship:

[0060]

[0061] in, is the azimuth error of the vehicle-mounted satellite positioning device, is the measured azimuth of the vehicle-mounted satellite positioning device, To simulate the standard azimuth.

[0062] Then, based on the current scene characteristics and device operating status, a dynamic threshold is obtained based on the dynamic signal fitting model. For example, in urban canyon scenarios, the dynamic threshold of the carrier-to-noise ratio is lower than in open scenarios to adapt to signal attenuation characteristics; in high-speed scenarios, the dynamic threshold range of the frequency shift change rate is expanded to avoid misjudging normal dynamics. The multi-dimensional anomaly detection engine simultaneously monitors three types of indicators: the signal layer detects anomalies such as sudden drops in the carrier-to-noise ratio; the geometric layer analyzes the distribution of positioning residuals; and the physical layer verifies the carrier kinematic constraints. The detection results are presented in the form of anomaly severity scores, and fuzzy logic is used to integrate the detection results of each dimension into quantitative indicators as multi-dimensional anomaly detection results.

[0063] Finally, collaborative verification incorporates a confidence-weighting mechanism, assigning weights based on the confidence level of each indicator. Error values ​​are integrated with multi-dimensional anomaly detection results and interactively verified to produce preliminary verification results. These results are presented in a three-dimensional visualization, including a positioning error heat map, anomaly event timeline, and device health status. The preliminary verification report includes four core conclusions: positioning accuracy level, functional compliance determination, anomaly event list, and maintenance recommendations. This improves the quality assessment capabilities of vehicle-mounted satellite positioning devices.

[0064] S3. Construct a verification chain and an evidence storage chain, and combine the verification chain and the evidence storage chain to obtain a credible verification result of the preliminary verification result.

[0065] In this embodiment, the verification chain is constructed by first collecting raw observation data streams from the positioning device in real time through a multi-source sensor network on the vehicle terminal. These data streams include, but are not limited to, raw satellite observations, inertial measurement unit data, wheel speedometer pulse signals, and environmental images captured by cameras. These raw observation data streams are timestamped and arranged in strict chronological order to form an unalterable time-series evidence chain. To ensure the integrity and robustness of the time-series evidence chain, an improved chain-based sharding storage technology is employed to segment the continuous data stream into fixed-length data blocks. Each block generates a unique hash value and is chain-bound to the hash value of the previous block. The sharding nodes of the time-series evidence chain are deployed on the vehicle device, roadside base stations, and cloud servers, forming a three-tiered storage architecture. In the distributed verification node deployment phase, roadside base stations with edge computing capabilities along the road are selected as core nodes, and adjacent vehicle terminals serve as lightweight nodes to construct a heterogeneous verification network. Each node loads a lightweight consensus algorithm and performs parallel verification on the received evidence chain shards: the core nodes perform full data verification, while the lightweight nodes only verify key feature parameters. When the nodes reach a verification consensus, the evidence chain segment is marked as a trusted segment. Ultimately, all trusted segments are spliced ​​together in chronological order to form a verification chain.

[0066] During the pre-verification phase of the preliminary verification results, the verification chain ensures the credibility of the results through a three-tiered verification mechanism. The first tier is temporal consistency verification, which aligns the positioning trajectory in the preliminary verification results with the original observation evidence chain stored in the verification chain in both time and space. A dynamic time warping algorithm is used to calculate trajectory similarity. When the similarity falls below a preset threshold, an anomaly flag is triggered and the confidence level of the result is reduced. The second tier is device status verification, which extracts the operating status parameters of the positioning device from the verification chain and cross-validates them against the device health assessment report in the preliminary verification results. If a sudden drop in the number of satellite signal locks and an abnormal decrease in the carrier-to-noise ratio are detected, the device status assessment is considered to be biased. The third tier is environmental adaptability verification, which dynamically adjusts the expected positioning accuracy threshold based on the environmental perception data in the verification chain. The pre-verification ultimately generates a first verification result containing a verification pass rate, a list of anomaly flags, and a confidence score, providing trusted input for subsequent evidence verification.

[0067] In this embodiment, an evidence chain is constructed. First, the original observation data stream is preprocessed, and dynamic fuzzification processing is implemented on sensitive information involving personal privacy. A multimodal data chain is established based on the time-series evidence chain. In the design of the privacy protection mechanism, a privacy protection middleware based on homomorphic encryption is deployed, and key data such as positioning trajectories are encrypted and stored. The consensus mechanism of the evidence chain only allows authorized institutions to participate in the consensus process to ensure the credibility of the evidence operation. The multimodal evidence chain finally generated includes a triple structure of data fingerprint chain, operation audit chain, and device identity chain.

[0068] During the evidence verification phase, multimodal data segments associated with the first verification result are first extracted from the evidence chain, including raw observations from the positioning device, environmental perception data, and device work logs. The evidence verification process employs a three-tiered verification architecture: the first tier is data integrity verification, which performs tamper detection by comparing the hash value of the data block to be verified with the hash fingerprint stored in the evidence chain. The second tier is privacy protection compliance verification, using formal verification tools to check whether face blurring meets anonymization standards and whether location data truncation retains sufficient accuracy. The third tier is business logic consistency verification, comparing key parameters from the first verification result with historical data in the evidence chain for trends, and using long-short-term memory networks to detect abnormal mutation points. The evidence verification ultimately generates a second verification result containing the verification pass rate, privacy compliance score, and data tampering suspicion index. This provides trusted input for subsequent verification and significantly enhances the auditability of the verification results.

[0069] Furthermore, during the post-verification phase, the verification chain performs a final verification of the second verification result through a three-layer verification mechanism. First, a deep verification of temporal consistency is performed, aligning the key parameters in the second verification result with the original temporal evidence chain stored in the verification chain in both time and space. A dynamic time warping algorithm is used to calculate trajectory similarity. In the event of an anomaly, an anomaly flag is triggered and a manual review process is initiated. Second, device status traceability verification is implemented, extracting the positioning device's operating status parameters from the verification chain and cross-validating them with the second verification result. The probability of device failure is calculated using a Bayesian network. Finally, a final verification of environmental adaptability is performed, dynamically adjusting the expected positioning accuracy threshold based on the high-precision map data in the verification chain. Post-verification ultimately generates a trusted verification result that includes a verification pass rate, anomaly root cause analysis, and a confidence matrix. Subsequent processing is automatically triggered through blockchain smart contracts. Through spatiotemporal coupling verification, device status traceability, and intelligent environmental adaptation, the credibility of the preliminary verification results is enhanced.

[0070] S4. Construct a cloud-based digital twin model based on the trusted verification result, and obtain a digital twin of the vehicle-mounted satellite positioning device according to the cloud-based digital twin model.

[0071] In this embodiment, a digital twin of a vehicle-mounted satellite positioning device is constructed. The trusted verification results are first standardized and subjected to deep feature extraction. Core parameters such as positioning error, signal-to-noise ratio, and device temperature are normalized to eliminate dimensional differences and ensure comparability of verification results across batches and scenarios. The feature extraction process deploys an improved random forest algorithm to screen key features from standardized data. To reveal nonlinear correlations between features, the mutual information method is used to calculate correlations between features. A weighted feature association network is constructed, with nodes representing key features and edge weights reflecting the strength of the correlation between features.

[0072] Furthermore, a graph convolutional network is introduced to perform topological analysis on the feature association network, identifying deep correlation patterns such as positioning error propagation paths and device fault coupling patterns. The resulting feature association spectrum contains triple information: feature importance ranking, association path topology, and dynamic influence coefficients. This provides a precise feature association map for the cloud-based digital twin model, enabling it to accurately simulate the complex behavioral patterns of on-board satellite positioning devices in the real world.

[0073] In this embodiment, the cloud-based digital twin model adopts a three-dimensional layered architecture, consisting of a physical layer, a signal layer, and a behavioral layer, and realizes data interaction and model coupling through a characteristic correlation spectrum. The physical layer focuses on the hardware characteristics of the vehicle-mounted satellite positioning device, and the modeling scope covers core parameters such as antenna phase center deviation, crystal oscillator frequency drift, and RF front-end noise coefficient. The performance evolution law of the vehicle-mounted satellite positioning device under temperature changes, mechanical vibrations, and electromagnetic interference is established through the finite element analysis method. The behavioral layer uses the Markov decision process to model the dynamic response of the device in complex scenarios. The state transition probability matrix is ​​dynamically updated by the error coupling coefficient in the characteristic correlation spectrum. The Kalman filter is used as the core fusion algorithm. Its state equation and observation equation dynamically adjust the noise covariance matrix through the error propagation path in the characteristic correlation spectrum to ensure a high degree of consistency between the model output and the physical entity state.

[0074] Furthermore, the hybrid modeling process combines the advantages of data-driven and mechanism-based modeling. First, key features in the feature correlation spectrum are mapped to each model layer: physical layer parameters are associated with hardware test data through support vector regression; the signal layer noise model uses a generative adversarial network to fit the measured signal distribution and generate interference signal samples that conform to real-world scenarios; and the behavioral layer decision model, combined with a deep reinforcement learning algorithm, dynamically adjusts action strategies based on scenario characteristics in the feature correlation spectrum. Model coupling is achieved through a bidirectional mapping mechanism. During forward mapping, physical layer parameters drive the signal layer to generate simulated observation data, and the behavioral layer predicts positioning trajectories based on this simulated data. During reverse mapping, measured positioning errors are used to invert the cause of hardware failures using the adjoint method. The resulting cloud-based digital twin model supports bidirectional closed-loop verification: forward simulation can predict positioning performance boundaries in different scenarios, and reverse simulation can trace hardware failures, improving the maintainability of vehicle-mounted satellite positioning devices.

[0075] The cloud-based digital twin model includes:

[0076] First, based on multi-body dynamics and signal propagation theory, the dynamic behavior of the vehicle-mounted satellite positioning device is described, and the state evolution equation is constructed to satisfy the following relationship:

[0077]

[0078] in, is the state vector, is the evolution function, To influence the parameters, is the physical parameter, For noise.

[0079] Secondly, establish the observation equation to satisfy the following relationship:

[0080]

[0081] in, is the observation vector, is the observation function, is the state vector, is the observation parameter, is the observation noise.

[0082] Finally, combined with the observation equation, the state of the twin is updated based on the real-time verification data to satisfy the following relationship:

[0083]

[0084] in, is the updated state vector, is the state vector, is the change in the state vector.

[0085] In this embodiment, a cloud-based digital twin model is used to achieve real-time dynamic mapping of vehicle-mounted satellite positioning devices, establishing a data-driven closed-loop mapping mechanism. First, a two-way communication link is established with the vehicle terminal via the Internet of Things protocol. Edge computing nodes are used to preprocess the raw positioning data, including coordinate system conversion, multi-source sensor data fusion, and noise filtering, to ensure the quality of data input to the cloud-based model. The cloud-based model deploys a lightweight neural network architecture and continuously receives real-time positioning streaming data uploaded by the vehicle terminal through a federated learning mechanism. The model parameters are dynamically adjusted using a spatiotemporal attention mechanism to synchronize the states of the physical entity and the digital twin.

[0086] During the dynamic mapping process, a 3D visualization engine for the digital twin was constructed, converting longitude and latitude coordinates into 3D spatial positions. The vehicle's dynamics model was then combined to render the motion trajectory in real time. Simultaneously, a state monitoring module for the digital twin was developed, employing a Kalman filter algorithm to compensate for key parameters such as satellite signal strength and positioning precision factor. When positioning drift or signal shielding anomalies were detected, the digital twin's warning status indicator was triggered.

[0087] A feedback optimization channel is established between the digital twin and the physical entity, transmitting cloud-based analysis results back to the vehicle terminal via digital thread technology. When the digital twin predicts a degradation in satellite signal quality, it automatically triggers the vehicle terminal to switch to inertial navigation mode or activate the auxiliary positioning module. A reinforcement learning algorithm is used in the continuous optimization phase to dynamically adjust model weight parameters based on the deviation between actual positioning results and the digital twin's predictions, forming an iterative closed loop.

[0088] S5. Establish a detection and verification platform for the vehicle-mounted satellite positioning device, detect the digital twin based on the detection and verification platform to obtain an intelligent verification result, and realize intelligent verification of the vehicle-mounted satellite positioning device.

[0089] In this embodiment, the inspection and calibration platform for the on-board satellite positioning devices of operating vehicles aims to establish a unified and fully integrated application platform. This platform is based on the production network, office network, and Internet network, and uses cloud computing technology and combines it with the Internet of Things technology platform. With the support of policies, regulations, various security assurance systems, and standard specification systems, it provides a comprehensive, coordinated, penetrating management information system that can support adaptive adjustment requirements.

[0090] See Figure 2 The figure shows the platform architecture diagram of the detection and verification platform; it includes the terminal layer: responsible for the access, data analysis and storage of satellite positioning devices, the storage layer: the business layer is mainly responsible for the maintenance and management of the basic data of the verification platform, the service layer: business services for each business object (such as users, standard instruments, etc.), the application layer: system applications based on various service combinations, and the interface layer: interface services between this system and external systems.

[0091] Specifically, the main functions of the testing and verification platform include but are not limited to vehicle satellite positioning device access services, positioning device data analysis services, positioning device control services, positioning device data storage services, positioning device data query services, testing and verification management, testing and verification report management, report statistics, system management, and handheld mobile terminal testing equipment.

[0092] See Figure 3 The figure shows the operation flow chart of the detection and verification platform; the operation process includes:

[0093] When the vehicle arrives at the inspection area, the owner or driver submits an application for inspection of the vehicle-mounted satellite positioning device to the inspection personnel. The inspection and inspection personnel use handheld mobile positioning device detection equipment to identify the vehicle license plate and form an inspection and inspection business entrustment agreement form. The content of the agreement form shall at least include: license plate number, positioning device manufacturer, positioning device model, positioning device identification, owner, contact information, email address, etc. The system automatically supplements the necessary information such as the inspection and inspection time, and supports printing in the specified format. The owner or driver can sign and confirm the agreement form.

[0094] After the handheld mobile terminal detection equipment identifies the vehicle license plate, it sends the license plate and other related information to the background inspection and management system. The inspection and management system establishes a temporary inspection and communication transmission channel between the operating vehicle satellite positioning device and the inspection and management system through the dynamic safety supervision platform of the transport vehicle and the operation service provider platform. It automatically makes relevant settings for the vehicle's satellite positioning device, temporarily changes the data sending interval to once every 5 seconds, and informs the on-site inspection personnel of the setting results through the inspection and management system. It starts the Global Navigation Satellite System (GNSS) standard signal, and restarts the satellite positioning device through the remote control of the handheld mobile terminal detection equipment.

[0095] After receiving normal feedback information from the handheld mobile terminal testing equipment, the on-site inspection personnel begin to conduct inspection and verification, including automatic inspection and manual inspection.

[0096] Automatic calibration: Based on the signal sent by the GNSS standard signal simulator, all satellite positioning data reported by the positioning device during the signal source time period are extracted, and the standard algorithm is used to calculate the error value between the actual data of the positioning device and the standard signal data, as well as whether the key functions meet the requirements, mainly including: positioning error: error value and whether it is within the allowable error range; speed error: error value and whether it is within the allowable error range; mileage error: error value and whether it is within the allowable error range; azimuth error: error value and whether it is within the allowable error range; overspeed alarm function: whether the alarm is generated as required.

[0097] The positioning point, speed, mileage, and azimuth of the detection and verification can all be set according to user needs, and the required analog signal source can be produced according to the needs. At the same time, the relevant parameters are configured on the platform, and the corresponding data source is selected for detection and verification.

[0098] Manual inspection: For parts of the inspection and verification process that cannot be automatically inspected, the inspection and verification platform supports manual inspection and the recording of test results. Inspection involves manually performing specified operations and then using the positioning device data query service or observing the positioning device status to conduct judgment and analysis. Inspection items mainly include: device appearance, left turn signal status, right turn signal status, brake status, low beam status, high beam status, and Adaptive Cruise Control (ACC) status.

[0099] The management system collects data on various measurement parameters according to the calibration requirements, and compares, analyzes and calculates the uploaded test data with the standard data on the platform to obtain the calibration results (qualified or unqualified).

[0100] The verification management system generates the verification results and transmits them back to the on-site verification personnel's handheld mobile terminal testing equipment, automatically generating a verification result feedback sheet. If the result is qualified, the original verification record and verification certificate will be automatically generated, and the qualified verification result will be returned to the on-site verification personnel's terminal. If the verification result is unqualified, the unqualified verification result and the actual verification data of the corresponding unqualified parameters will be returned to the on-site verification personnel's handheld mobile terminal testing equipment. The on-site verification personnel will notify the owner or driver of the vehicle submitted for inspection on the spot and sign and confirm the verification result on the verification result feedback sheet.

[0101] After the equipment calibration is completed, the GNSS standard signal source is turned off, and the system automatically restores the set equipment, resets it to its original state, and restarts the equipment. At the same time, it notifies the on-site inspection personnel through the mobile terminal.

[0102] After the entire inspection and calibration process is completed, the original calibration records and calibration certificates will be generated, and the calibration results can also be uploaded to relevant systems such as the provincial transportation administration system.

[0103] In this embodiment, a deep fault injection analysis is performed on the digital twin through a cloud-based digital twin model. Utilizing a historical fault database and real-time operation data, a long short-term memory network is used to perform time series modeling on the state evolution trajectory of the digital twin, identifying typical fault modes such as satellite signal loss of lock, multipath interference, and clock drift. For each type of fault mode, a fault propagation tree based on the Monte Carlo method is constructed to simulate the evolutionary path from a single fault point to a system-level failure, generating a sequence signal containing dimensional information such as the time of fault occurrence, duration, and parameter offset. This sequence signal uses a time-event dual-dimensional encoding method to convert physical quantities such as satellite signal carrier-to-noise ratio mutations and positioning solution divergence into standardized fault feature vectors. To enhance signal generalization, a generative adversarial network is introduced to perform data enhancement on real fault samples, generating a virtual fault case library covering complex scenarios such as extreme weather and urban canyons, providing high-fidelity training data for subsequent positioning behavior fitting.

[0104] Furthermore, typical fault sequence signals are injected into the satellite navigation signal simulation source to generate radio frequency signals containing specific fault characteristics, and at the same time, closed-loop interaction is carried out with the digital twin. An adaptive Kalman filter algorithm is used to perform spatiotemporal registration of multimodal signals to solve the data synchronization problem caused by differences in sampling rates of different sensors. In the signal fusion process, a transfer learning mechanism is introduced, and a pre-trained deep residual network is used to extract deep signal features. The distribution differences between the measured signal and the simulated signal are eliminated through domain adaptation technology. Based on the fused signal, measurement parameters such as position, speed, and attitude are calculated in real time. In order to improve parameter accuracy, a dynamic weight adjustment strategy is developed to automatically assign confidence weights to each modal signal according to the signal quality indicator to ensure that centimeter-level accuracy can be maintained even in complex electromagnetic environments.

[0105] In an optional embodiment, the detection and verification platform adopts a dual-channel parallel verification architecture to establish a standard metrology parameter baseline library and a dynamic threshold adaptation mechanism. In the standard channel, the baseline trajectory data of the vehicle-mounted satellite positioning device is obtained by a high-precision laser tracker. In the test channel, the metrology parameters output by the digital twin are compared with the baseline data in real time, and the parameter deviation is calculated using the Mahalanobis distance algorithm, combined with the support vector machine classifier to achieve intelligent identification of abnormal states. In order to solve the problem of high missed detection rate of traditional threshold methods, the normal parameter fluctuation range is automatically updated based on historical verification data. When a parameter abnormality is detected, the reverse tracing function of the digital twin is triggered to locate the root cause of the fault and generate a detection and verification report containing the fault type, impact assessment, repair suggestions, etc. The final intelligent verification result not only includes quantitative evaluation indicators, but also displays the fault propagation path and parameter evolution trend through the digital twin visualization interface, providing decision support for equipment maintenance.

[0106] See Figure 4 In an optional embodiment, the present invention provides an intelligent verification system for a vehicle-mounted satellite positioning device. The system includes an input device, an output device, a processor, and a memory, wherein the hardware components are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions and execute the specific implementation steps of the intelligent verification method for a vehicle-mounted satellite positioning device provided by the present invention. The intelligent verification system for a vehicle-mounted satellite positioning device provided by the present invention has a complete structure, objective stability, and enhances the overall applicability and practical application capabilities of the present invention.

[0107] In summary, the method of the present invention provides an intelligent verification method and system for a vehicle-mounted satellite positioning device, which improves positioning accuracy and environmental adaptability through multimodal signal fusion and dynamic modeling, and the collaborative verification mechanism enhances the comprehensiveness and accuracy of data verification; combines the verification chain and the evidence chain to ensure that the verification results are tamper-proof and traceable, and builds a highly reliable data foundation; the digital twin realizes real-time mapping of the device's operating status, reducing the cost of physical testing; the detection and verification platform automatically completes the verification, forming a closed process loop, significantly improving the verification efficiency and device reliability, and providing full life cycle intelligent protection for vehicle-mounted satellite positioning devices. The method of the present invention is easy to understand, simple to calculate, with a small workload, and convenient for engineering application, providing a theoretical basis and technical support for the further development of intelligent traffic management technology.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. An intelligent calibration method for a vehicle-mounted satellite positioning device, characterized in that: The steps include: Combining multimodal simulation signals and adaptive scene signals to build a dynamic signal fitting model for vehicle-mounted satellite positioning devices; Performing collaborative verification on the vehicle-mounted satellite positioning device according to the dynamic signal fitting model to obtain a preliminary verification result; Constructing a verification chain and an evidence storage chain, and combining the verification chain and the evidence storage chain to obtain a credible verification result of the preliminary verification result; Constructing a cloud-based digital twin model based on the trustworthy verification result, and obtaining a digital twin of the vehicle-mounted satellite positioning device according to the cloud-based digital twin model; Establishing a detection and verification platform for the vehicle-mounted satellite positioning device, detecting the digital twin based on the detection and verification platform to obtain intelligent verification results, thereby realizing intelligent verification of the vehicle-mounted satellite positioning device; The collaborative verification of the vehicle-mounted satellite positioning device according to the dynamic signal fitting model to obtain a preliminary verification result includes: Obtaining a dynamic signal through the dynamic signal fitting model, and fusing the dynamic signal with the original data of the vehicle-mounted satellite positioning device to obtain a fused signal; Filtering the fused signal to obtain a corrected positioning trajectory of the vehicle-mounted satellite positioning device; Obtaining a standard value of the dynamic signal, and obtaining an error value of the corrected positioning trajectory in combination with the standard value; Setting a dynamic threshold according to the dynamic fitting model, and obtaining a multi-dimensional anomaly detection result of the vehicle-mounted satellite positioning device based on the dynamic threshold; The error value and the multi-dimensional anomaly detection result are combined to perform collaborative verification to obtain the preliminary verification result.

2. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 1, characterized in that: The method of combining the multimodal simulation signal and the adaptive scene signal to construct a dynamic signal fitting model for the vehicle-mounted satellite positioning device includes: Performing virtual scene fitting based on a simulation signal generating device to obtain the multimodal simulation signal; Constructing a three-dimensional road model, and mapping the physical environment according to the three-dimensional road model to obtain the adaptive scene signal; The multimodal simulation signal and the adaptive scene signal are dynamically fused to construct the dynamic signal fitting model.

3. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 1, characterized in that: The constructing of the verification chain and the evidence storage chain, and combining the verification chain and the evidence storage chain to obtain a credible verification result of the preliminary verification result, includes: Obtaining a time series evidence chain, and constructing the verification chain based on the time series evidence chain and distributed verification nodes; Establish a multimodal data chain, and build the evidence chain based on the multimodal data chain in combination with a privacy protection mechanism; The preliminary verification result is cross-checked in combination with the verification chain and the evidence storage chain to obtain the credible verification result.

4. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 3, characterized in that: The cross-verification of the preliminary verification result by combining the verification chain and the evidence storage chain to obtain the credible verification result includes: Performing pre-verification on the preliminary verification result based on the verification chain to obtain a first verification result; Performing evidence verification on the first verification result according to the evidence chain to obtain a second verification result; The second verification result is post-verified according to the verification chain to obtain the trusted verification result.

5. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 1, characterized in that: The step of constructing a cloud-based digital twin model based on the trustworthy verification result and obtaining a digital twin of the vehicle-mounted satellite positioning device according to the cloud-based digital twin model includes: Performing standardization and feature extraction on the credibility verification results to obtain key features, and constructing a feature correlation spectrum based on the key features; Constructing a cloud-based digital twin model of the vehicle-mounted satellite positioning device according to the feature correlation spectrum; The digital twin is obtained by performing real-time dynamic mapping of the vehicle-mounted satellite positioning device based on the cloud digital twin model.

6. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 5, characterized in that: The step of constructing a cloud-based digital twin model of the vehicle-mounted satellite positioning device according to the feature correlation spectrum includes: Establishing a model layered architecture of the cloud-based digital twin model, wherein the model layered architecture includes a physical layer, a signal layer, and a behavior layer; Based on the feature association spectrum and combined with the model hierarchical architecture, hybrid modeling is performed on the vehicle-mounted satellite positioning device to obtain the cloud-based digital twin model.

7. The intelligent calibration method of the vehicle-mounted satellite positioning device according to claim 1, characterized in that: The establishment of a detection and verification platform for the vehicle-mounted satellite positioning device includes: Establishing a multi-level platform architecture, comprising a terminal layer, a storage layer, a service layer, an application layer, and an interface layer; The detection and verification platform is obtained based on the multi-level platform architecture.

8. The intelligent calibration method for a vehicle-mounted satellite positioning device according to claim 1, characterized in that: The method of detecting the digital twin based on the detection and verification platform to obtain an intelligent verification result and realize intelligent verification of the vehicle-mounted satellite positioning device includes: Obtaining a typical failure mode of the digital twin based on a cloud-based digital twin model, and acquiring a sequence signal of the typical failure mode; Fitting the positioning behavior of the vehicle-mounted satellite positioning device by combining the multimodal simulation signal and the sequence signal, and obtaining the metrological parameters of the vehicle-mounted satellite positioning device; The measurement parameters are compared and verified by the detection and verification platform to obtain a detection and verification report as the intelligent verification result of the vehicle-mounted satellite positioning device.

9. An intelligent calibration system for a vehicle-mounted satellite positioning device, characterized in that: The system includes an input device, an output device, a processor and a memory, wherein the input device, the output device, the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the intelligent verification method of the vehicle-mounted satellite positioning device according to any one of claims 1 to 8.

Images