Positioning strategy adjustment method and device based on UPD time sequence stability evaluation, electronic equipment and storage medium

Abstract

The invention discloses a positioning strategy adjustment method and device based on UPD time sequence stability evaluation, electronic equipment and a storage medium, and belongs to the field of satellite navigation high-precision positioning, and the method comprises the steps: obtaining multi-source time sequence observation data of a target satellite in a real-time monitoring time period, and constructing a corresponding time sequence feature vector sequence; extracting a characteristic component sequence corresponding to a UPD resolving residual error, respectively calculating a first statistical dispersion degree in a short time window and a second statistical dispersion degree in a long time window, and taking a ratio of the first statistical dispersion degree to the second statistical dispersion degree as an instability index; generating a comprehensive reliability evaluation result in combination with the time sequence feature vector sequence and the instability index; and comparing the comprehensive reliability evaluation result with a preset reference, and executing a corresponding positioning strategy adjustment operation according to the result. By implementing the method and the device, the problem of difficulty in accurately identifying tiny mutation or misjudging a high-noise environment in a complex environment in the prior art can be solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of satellite navigation high-precision positioning technology, and particularly relates to a positioning strategy adjustment method and device based on UPD time sequence stability evaluation, electronic equipment and storage medium. BACKGROUND

[0002] In high-precision satellite navigation positioning services, the accurate estimation of phase fractional offset (UPD) is the key to realize the rapid fixing of integer ambiguity. Due to the influence of ionosphere activity, multipath effect, hardware thermal noise and other factors, UPD data often presents nonlinear time-varying fluctuations. Therefore, accurately evaluating the time sequence stability of UPD is of great significance to maintain the usability and robustness of high-precision positioning services.

[0003] However, the existing UPD stability evaluation method usually relies on a single instantaneous residual threshold or a single fixed window statistical variance for judgment. The main defect of this method is the lack of relative quantization ability of signal fluctuation components, which makes it difficult to effectively distinguish whether the data fluctuation is caused by the overall cumulative increase of environmental background noise or by the sudden abnormality of the signal itself (such as a small cycle slip). In actual application, this "absolute threshold" discrimination method has obvious drawbacks: in a bad observation environment, the available signal is often misjudged as abnormal (false alarm) due to the overall large background noise; while in a stable environment but with a small signal mutation, it fails to break through the absolute threshold, resulting in missed detection (false negative), which seriously affects the reliability of positioning solution. SUMMARY

[0004] The embodiments of the present application provide a positioning strategy adjustment method and device based on UPD time sequence stability evaluation, electronic equipment and storage medium, which can solve the problem that the single threshold evaluation in the prior art cannot accurately identify small mutations or misjudge high-noise environments in complex environments.

[0005] An embodiment of the present application provides a positioning strategy adjustment method based on UPD time sequence stability evaluation, comprising: Obtaining multi-source time sequence observation data of a target satellite in a real-time monitoring period; wherein the multi-source time sequence observation data includes UPD solution residual values; According to the multi-source time sequence observation data, a time sequence feature vector sequence is constructed; wherein each sequence element in the time sequence feature vector sequence is a feature vector corresponding to an epoch, and each dimension component of the feature vector corresponds to a kind of multi-source time sequence observation data; extracting a feature component sequence corresponding to the UPD calculation residual error value in the time sequence feature vector sequence, respectively calculating a first statistical dispersion of the feature component sequence within a preset short time window and a second statistical dispersion of the feature component sequence within a preset long time window; calculating a ratio of the first statistical dispersion and the second statistical dispersion, and determining the ratio as an instability indicator; generating a comprehensive reliability evaluation result of UPD data in combination with the time sequence feature vector sequence and the instability indicator; wherein the comprehensive reliability evaluation result includes an availability score of UPD data at a current epoch and a prediction of a future effective time length of UPD data; comparing the comprehensive reliability evaluation result with a preset reference, and performing a positioning strategy adjustment operation according to a matching result.

[0006] Further, the multi-source time sequence observation data further includes a carrier-to-noise ratio value and a multipath effect value. The constructing a time sequence feature vector sequence according to the multi-source time sequence observation data includes: For each observation epoch included in the multi-source time sequence observation data, extracting a UPD calculation residual error value, a carrier-to-noise ratio value and a multipath effect value corresponding to the current observation epoch; respectively normalizing the UPD calculation residual error value, the carrier-to-noise ratio value and the multipath effect value corresponding to the current observation epoch to obtain a normalized residual error component, a normalized carrier-to-noise ratio component and a normalized multipath component of the current observation epoch; splicing the normalized residual error component, the normalized carrier-to-noise ratio component and the normalized multipath component of the current observation epoch to generate a feature vector of the current observation epoch; sequentially arranging the generated feature vectors corresponding to multiple observation epochs in the order of observation time to construct a time sequence feature vector sequence.

[0007] Further, the extracting a feature component sequence corresponding to the UPD calculation residual error value in the time sequence feature vector sequence, respectively calculating a first statistical dispersion of the feature component sequence within a preset short time window and a second statistical dispersion of the feature component sequence within a preset long time window includes: extracting dimension data corresponding to the normalized residual error component from each feature vector included in the time sequence feature vector sequence to form a single-dimensional residual error feature sequence in the order of epochs; taking the current observation epoch as a cutoff time, in the residual error feature sequence, backtracking to the front to intercept a first data segment with a time span corresponding to a preset short time window, and a second data segment with a time span corresponding to a preset long time window; Calculate the standard deviation of all normalized residual components in the first data segment, and determine the calculated standard deviation as the first statistical dispersion; Calculate the standard deviation of all normalized residual components in the second data segment, and determine the calculated standard deviation as the second statistical dispersion.

[0008] Further, the combination of the time sequence feature vector sequence and the instability index generates a comprehensive reliability evaluation result of UPD data, including: Extract the feature vector corresponding to the current observation epoch from the time sequence feature vector sequence, and perform weighted fusion on the feature vector of the current observation epoch and the instability index by using a preset weighting coefficient to obtain a freshness score representing real-time quality of data; Trend fitting is performed on the time sequence feature vector sequence, the trend change rate of the fitting curve is calculated, and an integrity score representing the evolution trend of data is generated according to the trend change rate; Based on a preset reliability decay model, the freshness score and the integrity score are used as input parameters to calculate the dynamic validity period of UPD data; The freshness score, the integrity score, and the dynamic validity period of the UPD data are combined as a comprehensive reliability evaluation result.

[0009] Further, the trend fitting of the time sequence feature vector sequence, the calculation of the trend change rate of the fitting curve, and the generation of the integrity score representing the evolution trend of data according to the trend change rate, include: Extract the time sequence data corresponding to the normalized residual component from the time sequence feature vector sequence to construct a normalized residual component sequence; A polynomial fitting is performed on the normalized residual component sequence by using the least square method to obtain a fitting curve function capable of representing the trend of residual change; Calculate the first derivative value of the fitting curve function at the time corresponding to the current observation epoch, and determine the absolute value of the first derivative value as the trend change rate; The trend change rate is converted into an integrity score by using a preset negative correlation mapping function; the larger the value of the trend change rate, the lower the integrity score.

[0010] Further, the reliability decay model includes a linear correction function and an exponential decay function; Based on the preset reliability decay model, the freshness score and the integrity score are used as input parameters to calculate the dynamic validity period of the UPD data, including: Substitute the freshness score into the linear correction function to calculate a time correction coefficient; substitute the integrity score into the exponential decay function to calculate a decay factor; a product of the preset reference validity duration, the time correction coefficient and the decay factor is calculated, and the calculated product is determined as the dynamic validity period of the UPD data.

[0011] Further, the comparison of the comprehensive reliability evaluation result with the preset reference and the execution of the matching positioning strategy adjustment operation according to the comparison result include: The freshness score is compared with a preset first quality threshold and a preset minimum available threshold respectively, the integrity score is compared with a preset second quality threshold, and the dynamic validity period is compared with a preset time threshold; If the freshness score is between the minimum available threshold and the first quality threshold, or the integrity score is lower than the second quality threshold, an operation of reducing the weight of the target satellite in the positioning solution model is executed; If the dynamic validity period is lower than the time threshold, an operation of requesting and prefetching standby UPD data is executed; If the freshness score is lower than the minimum available threshold, at least one of an operation of eliminating the target satellite and an operation of switching data sources is executed; wherein the minimum available threshold is less than the first quality threshold.

[0012] On the basis of the above-mentioned method embodiment, the application provides a device embodiment.

[0013] An embodiment of the application provides a positioning strategy adjustment device based on UPD time sequence stability evaluation, which comprises a data acquisition module, a feature construction module, an index calculation module, a comprehensive evaluation module and a strategy adjustment module. The data acquisition module is used for acquiring multi-source time sequence observation data of a target satellite in a real-time monitoring period; wherein the multi-source time sequence observation data comprises a UPD solution residual value. The feature construction module is used for constructing a time sequence feature vector sequence according to the multi-source time sequence observation data; wherein each sequence element in the time sequence feature vector sequence is a feature vector corresponding to an ephemeris, and each dimension component of the feature vector corresponds to one kind of multi-source time sequence observation data. The index calculation module is used for extracting a feature component sequence corresponding to the UPD solution residual value in the time sequence feature vector sequence, calculating a first statistical dispersion of the feature component sequence in a preset short time window and a second statistical dispersion of the feature component sequence in a preset long time window respectively, calculating a ratio of the first statistical dispersion to the second statistical dispersion, and determining the ratio as an instability index. The comprehensive evaluation module is used to combine the time-series feature vector sequence and the instability index to generate a comprehensive reliability evaluation result for UPD data; wherein, the comprehensive reliability evaluation result includes an availability score for UPD data in the current epoch and a prediction of the future effective duration of UPD data; The strategy adjustment module is used to compare the comprehensive reliability evaluation results with a preset benchmark, and perform a matching positioning strategy adjustment operation based on the comparison results.

[0014] Based on the above method embodiments, the present invention provides corresponding electronic device embodiments.

[0015] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the positioning strategy adjustment method based on UPD timing stability evaluation as described in any of the above-described method embodiments.

[0016] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments.

[0017] One embodiment of the present invention provides a storage medium storing a computer program thereon, wherein, when the computer program is running, it controls the device where the storage medium is located to execute any of the positioning strategy adjustment methods based on UPD timing stability evaluation described in the above-described method embodiments.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a positioning strategy adjustment method, apparatus, electronic device, and storage medium based on UPD temporal stability evaluation. The method acquires multi-source temporal observation data of a target satellite during a real-time monitoring period and constructs a corresponding temporal feature vector sequence. It extracts the feature component sequence corresponding to the UPD solution residuals from this data, calculates its first statistical dispersion within a short time window and its second statistical dispersion within a long time window, and uses the ratio of these two as an instability index. Combining the temporal feature vector sequence and the instability index, it generates a comprehensive reliability evaluation result for the UPD data, including an availability score for the current epoch and a prediction of future effective duration. The comprehensive reliability evaluation result is compared with a preset benchmark, and corresponding positioning strategy adjustment operations are performed accordingly.

[0019] This application extracts feature component sequences, calculates their first statistical dispersion within a preset short time window and their second statistical dispersion within a preset long time window, and uses the ratio of the two to construct an instability index, thereby realizing the relative quantification of the instantaneous fluctuation characteristics of the signal and the level of environmental background noise. This effectively distinguishes whether the fluctuation of the UPD solution residual is caused by the accumulation of environmental noise or by sudden signal anomalies, and solves the technical problem that existing single threshold evaluation is difficult to accurately identify small mutations or misjudge high-noise environments in complex environments. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a positioning strategy adjustment method based on UPD timing stability evaluation, provided by an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of a positioning strategy adjustment device based on UPD timing stability evaluation provided in an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, to address the problem that existing technologies using a single threshold evaluation method are insufficient to accurately identify minute mutations or misjudge high-noise environments in complex settings, an embodiment of the present invention provides a positioning strategy adjustment method based on UPD temporal stability evaluation, comprising at least the following steps: Step S1: Obtain multi-source time-series observation data of the target satellite during the real-time monitoring period; wherein, the multi-source time-series observation data includes UPD solution residual values; Specifically, in implementing the positioning strategy adjustment method based on UPD temporal stability evaluation, the first step is to acquire multi-source temporal observation data of the target satellite during the real-time monitoring period. The real-time monitoring period refers to the time window during which satellite signals are continuously tracked and data is collected at a preset sampling frequency; this period consists of multiple consecutive observation epochs. At each observation epoch, the receiving device or server performs data processing and status recording for each target satellite within its field of view, thus forming a dynamically updated data stream over time. The target satellite refers to a single satellite in the Global Navigation Satellite System currently participating in precise point positioning or real-time dynamic positioning calculations.

[0024] As the core input data of this method, the multi-source time-series observation data is not a single-dimensional data record, but a collection of various data types that can reflect the estimation quality of UPD (Uncalibrated Phase Delay). Specifically, the multi-source time-series observation data must include UPD solution residuals. These UPD solution residuals refer to the difference between the actual observed phase value in the observation equation and the theoretical phase value calculated based on the model when performing fractional phase bias estimation, or the posterior residual generated during the integer ambiguity fixing process. UPD solution residuals can intuitively reflect the degree of fit between the UPD estimation model and the observation data at the current epoch. When the observation environment is good and no cycle slips occur, the UPD solution residuals typically exhibit small random fluctuations near zero; however, when signal lock-out or environmental interference occurs, the UPD solution residuals will show significant shifts or oscillations.

[0025] To aid in assessing data stability from multiple perspectives, in a preferred embodiment, the multi-source time-series observation data further includes a carrier-to-noise ratio (CNR) and a multipath effect. The CNR represents the ratio of the received target satellite signal strength to the noise intensity; a higher CNR indicates better signal quality and lower noise levels in the observation data. The multipath effect refers to the multipath propagation error caused by reflections from surrounding objects during satellite signal transmission. The multipath effect reflects the complexity of the environment surrounding the station and the extent of reflection interference.

[0026] In acquiring the aforementioned data, for each observation epoch within the real-time monitoring period, the method synchronously collects and records the corresponding UPD solution residual value, carrier-to-noise ratio (CNR) value, and multipath effect value, and aligns and stores these different types of data in chronological order. This approach not only obtains a snapshot of the data at a single moment but, more importantly, preserves the evolutionary characteristics of the data over time, i.e., the "temporal" attribute. Using the UPD solution residual value, CNR value, and multipath effect value together as multi-source time-series observation data allows for the establishment of a multi-dimensional data view encompassing model fitting quality (residual), signal physical strength (CNR), and environmental interference level (multipath).

[0027] By acquiring the aforementioned multi-source time-series observation data containing multi-dimensional indicators, we can provide a rich and complete data foundation for the subsequent construction of high-dimensional time-series feature vector sequences, thereby ensuring that the subsequent stability evaluation is no longer limited to a single indicator, but is based on a comprehensive consideration of multi-source information.

[0028] Step S2: Construct a time series feature vector sequence based on the multi-source time series observation data; wherein, each sequence element in the time series feature vector sequence is a feature vector of the corresponding epoch, and each dimension of the feature vector corresponds to a type of multi-source time series observation data; In a preferred embodiment, the multi-source time-series observation data further includes: a carrier-to-noise ratio value and a multipath effect value; The step of constructing a time-series feature vector sequence based on the multi-source time-series observation data includes: For each observation epoch in the multi-source time-series observation data, extract the UPD solution residual value, carrier-to-noise ratio value and multipath effect value corresponding to the current observation epoch; The UPD solution residual value, carrier-to-noise ratio value and multipath effect value corresponding to the current observation epoch are normalized respectively to obtain the normalized residual component, normalized carrier-to-noise ratio component and normalized multipath component of the current observation epoch. The normalized residual component, normalized carrier-to-noise ratio component, and normalized multipath component of the current observation epoch are concatenated to generate the feature vector of the current observation epoch. The feature vectors corresponding to the multiple observation epochs are arranged sequentially according to the order of observation time to construct a time-series feature vector sequence.

[0029] Specifically, after acquiring multi-source time-series observation data, it is necessary to further construct a time-series feature vector sequence based on the multi-source time-series observation data. The time-series feature vector sequence is a set of mathematical descriptions used to characterize the evolution of the target satellite signal quality over time. Each element in the time-series feature vector sequence corresponds to an observation epoch within the real-time monitoring period, and each element is a multi-dimensional feature vector. Each dimension of the feature vector corresponds to a type of multi-source time-series observation data. In this way, observation indicators of different physical dimensions are mapped to a unified vector space to facilitate subsequent joint analysis.

[0030] In a preferred embodiment, to more comprehensively evaluate the reliability of the phase fractional deviation (UPD) data, the multi-source time-series observation data, in addition to including the UPD solution residual value, further includes the carrier-to-noise ratio (CNR) value and the multipath effect value. The CNR value reflects the physical strength of the satellite signal and the channel quality, while the multipath effect value reflects the level of environmental reflection interference around the station. The specific process of constructing the time-series feature vector sequence includes: for each observation epoch included in the real-time monitoring period, firstly, extracting the UPD solution residual value, CNR value, and multipath effect value corresponding to the current observation epoch. Since these three types of data have different physical dimensions and orders of magnitude, direct concatenation will cause the larger features to mask the smaller features; therefore, normalization processing is required for each.

[0031] The specific normalization process employs the Z-Score standardization method to eliminate the influence of dimensions and achieve data homogenization. For the current observation epoch, the real-time acquired data is standardized using the mean and standard deviation within a pre-stored historical statistical window.

[0032] The normalized calculation formula for the residual value of UPD is as follows: In the formula, This represents the normalized residual component of the current observation epoch; This represents the original UPD solution residual value extracted at the current observation epoch; This represents the statistical mean of the residual values ​​calculated by UPD within the preset historical window; This represents the statistical standard deviation of the residual values ​​calculated by UPD within the preset historical window.

[0033] The normalized calculation formula for the carrier-to-noise ratio is as follows: In the formula, This represents the normalized carrier-to-noise ratio component of the current observation epoch; This represents the raw carrier-to-noise ratio value extracted from the current observation epoch; This represents the statistical mean of the carrier-to-noise ratio values ​​within the preset historical window. This represents the statistical standard deviation of the carrier-to-noise ratio values ​​within the preset historical window.

[0034] The normalized calculation formula for multipath effect values ​​is as follows: In the formula, This represents the normalized multipath component of the current observation epoch; This represents the raw multipath effect value extracted from the current observation epoch; This represents the statistical mean of multipath effect values ​​within a preset historical window. This represents the statistical standard deviation of multipath effect values ​​within a preset historical window.

[0035] After obtaining the normalized residual component, normalized carrier-to-noise ratio component, and normalized multipath component of the current observation epoch, these components are concatenated according to a preset dimensional order to generate a feature vector characterizing the overall quality state of the current observation epoch. Subsequently, the generated feature vectors corresponding to each observation epoch are arranged sequentially according to the order of observation time, ultimately constructing a time-series feature vector sequence that reflects the dynamic changes in the data.

[0036] The time-series feature vector sequence constructed in the above manner not only eliminates the dimensional differences between different source data, enabling the model to equally handle residuals, signal strength, and environmental interference factors, but also preserves the trend information of data fluctuation over time through the serialization organization of the time dimension, providing standardized data input for subsequent accurate calculation of instability indicators and trend fitting.

[0037] Step S3: Extract the feature component sequence corresponding to the UPD solution residual value from the time-series feature vector sequence, calculate the first statistical dispersion of the feature component sequence within a preset short time window and the second statistical dispersion of the feature component sequence within a preset long time window; calculate the ratio of the first statistical dispersion to the second statistical dispersion, and determine the ratio as an instability index; In a preferred embodiment, the step of extracting the feature component sequence corresponding to the UPD solution residual value from the time-series feature vector sequence, and calculating the first statistical dispersion of the feature component sequence within a preset short time window and the second statistical dispersion of the feature component sequence within a preset long time window, includes: From each feature vector contained in the time-series feature vector sequence, extract the dimensional data corresponding to the normalized residual component, and form a single-dimensional residual feature sequence in epochal order; Using the current observation epoch as the cutoff time, backtrack from the residual feature sequence to extract a first data segment with a time span corresponding to a preset short time window and a second data segment with a time span corresponding to a preset long time window; Calculate the standard deviation of all normalized residual components in the first data segment, and determine the calculated standard deviation as the first statistical dispersion; Calculate the standard deviation of all normalized residual components in the second data segment, and determine the calculated standard deviation as the second statistical dispersion.

[0038] Specifically, after constructing the time-series feature vector sequence, in order to quantitatively evaluate the stability of the UPD data at the current moment, it is necessary to extract instability indicators from the time-series feature vector sequence. First, from each feature vector contained in the time-series feature vector sequence, the dimensional data corresponding to the normalized residual components are precisely extracted. Since the multi-source data has already been normalized in the previous process, the extracted data is now a pure numerical sequence free from the influence of dimensions. These values ​​are then recombine according to the chronological order of the observation epochs to form a one-dimensional residual feature sequence. This residual feature sequence records the fluctuation trajectory of the UPD solution residuals on the time axis and serves as the data foundation for subsequent statistical analysis.

[0039] Subsequently, to capture the difference between the instantaneous fluctuation characteristics of the signal and the background noise level, the method performs backward truncation in the residual feature sequence, using the current observation epoch as the cutoff point. This process involves two data windows with different time spans: a preset short-time window to capture the latest signal dynamics, and a preset long-time window to establish a baseline for environmental background noise. Specifically, a first data segment corresponding to the time span of the preset short-time window is truncated from the residual feature sequence, while a second data segment corresponding to the time span of the preset long-time window is truncated. Typically, the length of the preset long-time window is significantly longer than the length of the preset short-time window to ensure that the second data segment contains sufficient historical samples to reflect statistical regularities.

[0040] Next, statistical dispersion is calculated for the first data segment and the second data segment respectively. In this embodiment, statistical dispersion is specifically represented by the standard deviation of the data. The standard deviation of all normalized residual components in the first data segment is calculated, and the calculated standard deviation is determined as the first statistical dispersion, denoted as . The first statistical dispersion reflects the severity of the instantaneous jitter of the UPD residuals around the current time.

[0041] The formula for calculating the first statistical dispersion is as follows: In the formula, This represents the total number of sample points contained in the first data segment; Indicates the first data segment The normalized residual component values; This represents the arithmetic mean of all normalized residual components in the first data segment.

[0042] Simultaneously, the standard deviation of all normalized residual components in the second data segment is calculated, and the calculated standard deviation is determined as the second statistical dispersion, denoted as . The second statistical dispersion reflects the overall background noise level of the observation environment or the average fluctuation of UPD data over a longer period of time.

[0043] The formula for calculating the second statistical dispersion is as follows: In the formula, This represents the total number of sample points contained in the second data segment; Indicates the second data segment The normalized residual component values; This represents the arithmetic mean of all normalized residual components in the second data segment.

[0044] Finally, the ratio of the first statistical dispersion to the second statistical dispersion is calculated, and this ratio is determined as the instability index. The instability index is a relative quantitative value used to characterize the degree of deviation of the signal fluctuation at the current moment relative to historical background noise.

[0045] The formula for calculating the instability index is as follows: In the formula, This indicates the instability index.

[0046] When the instability index is significantly greater than 1, it means that the current instantaneous fluctuation amplitude far exceeds the historical background noise level, indicating that there may be a sudden signal loss or cycle slip interference; when the instability index is close to 1, it means that the current fluctuation is within the normal background noise range.

[0047] By calculating the ratio of the first statistical dispersion to the second statistical dispersion as an instability index, a relative evaluation of the stability of UPD data is achieved. This effectively shields against misjudgments caused by excessively large absolute residual values ​​in high-noise environments, while accurately capturing minute sudden anomalies in low-noise environments, significantly improving the sensitivity and robustness of anomaly detection in complex observation scenarios.

[0048] Step S4: Combine the time-series feature vector sequence with the instability index to generate a comprehensive reliability evaluation result for the UPD data; wherein, the comprehensive reliability evaluation result includes an availability score for the UPD data in the current epoch and a prediction of the future effective duration of the UPD data; In a preferred embodiment, the step of generating a comprehensive reliability evaluation result for UPD data by combining the time-series feature vector sequence and the instability index includes: The feature vector corresponding to the current observation epoch is extracted from the time-series feature vector sequence. The feature vector of the current observation epoch is weighted and fused with the instability index using a preset weighting coefficient to obtain a freshness score that characterizes the real-time quality of the data. The time-series feature vector sequence is subjected to trend fitting, the trend change rate of the fitted curve is calculated, and an integrity score representing the evolution trend of the data is generated based on the trend change rate. Based on a preset reliability decay model, the dynamic validity period of UPD data is calculated using the freshness score and the integrity score as input parameters. The freshness score, the integrity score, and the dynamic validity period of the UPD data are combined to form a comprehensive reliability evaluation result.

[0049] In a preferred embodiment, the step of performing trend fitting on the time-series feature vector sequence, calculating the trend change rate of the fitted curve, and generating an integrity score representing the evolution trend of the data based on the trend change rate includes: Extract the time series data corresponding to the normalized residual components from the time series feature vector sequence, and construct the normalized residual component sequence; The normalized residual component sequence is fitted with a polynomial using the least squares method to obtain a fitting curve function that can characterize the trend of residual change. Calculate the first derivative of the fitted curve function at the time corresponding to the current observation epoch, and determine the absolute value of the first derivative as the trend change rate; The trend change rate is converted into an integrity score using a preset negative correlation mapping function; wherein, the larger the value of the trend change rate, the lower the integrity score.

[0050] In a preferred embodiment, the reliability decay model includes a linear correction function and an exponential decay function; The dynamic validity period of the UPD data, calculated based on a preset reliability decay model and using the freshness score and the integrity score as input parameters, includes: Substitute the freshness score into the linear correction function to calculate the time correction coefficient; Substitute the integrity score into the exponential decay function to calculate the decay factor; Calculate the product of the preset baseline validity period, the time correction coefficient, and the attenuation factor, and determine the calculated product as the dynamic validity period of the UPD data.

[0051] Specifically, after obtaining the time-series feature vector sequence and instability index, it is necessary to further combine these two types of information to generate a comprehensive reliability evaluation result for the UPD data. This comprehensive reliability evaluation result is not a single-dimensional numerical value, but a set of multi-dimensional evaluation indicators, specifically including the usability score of the UPD data in the current epoch (i.e., freshness and completeness) and the prediction of the duration for which the UPD data will remain valid in the future (i.e., dynamic validity period). This process is achieved through a multi-level evaluation model, aiming to provide a comprehensive profile of the data from three aspects: real-time quality, evolutionary trend, and time dimension.

[0052] First, a freshness score representing the real-time quality of the data is calculated. A feature vector corresponding to the current observation epoch is extracted from the temporal feature vector sequence. This feature vector, constructed through previous steps, includes the normalized residual component, normalized carrier-to-noise ratio component, and normalized multipath component at the current time. The feature vector of the current observation epoch is then weighted and fused with the instability index calculated in the previous step using preset weighting coefficients. This weighted fusion process aims to comprehensively consider the signal's physical strength, environmental interference level, and statistical stability.

[0053] The specific formula for calculating the freshness score is as follows: In the formula, This indicates the freshness score; , , These represent the normalized residual component, normalized carrier-to-noise ratio component, and normalized multipath component of the current observation epoch, respectively. This indicates the instability index; , , , These are preset weighting coefficients corresponding to each indicator, and the sum of all weighting coefficients is 1. According to this formula, the smaller the residual, the higher the noise-carrying ratio, the smaller the multipath propagation, and the lower the instability index, the higher the calculated freshness score.

[0054] Secondly, the step of calculating the integrity score representing the evolution trend of the data is performed. The core of this step is to identify the trend of data change, that is, to determine whether the data is in a slowly deteriorating drift state. First, the time series data corresponding to the normalized residual components are extracted from the time series feature vector sequence to construct the normalized residual component sequence. Then, the least squares method is used to perform polynomial fitting on the normalized residual component sequence to construct a fitting curve function that can represent the law of residual change over time.

[0055] Let the fitted curve function be... Calculate the first derivative of the fitted curve function at the time corresponding to the current observation epoch. The absolute value of the first derivative is determined as the trend change rate.

[0056] In the formula, This represents the rate of change of the trend; This represents the current observation epoch. The trend change rate reflects the rate of change of the UPD residuals. Next, a preset negative correlation mapping function is used to convert the trend change rate into an integrity score. Since a larger trend change rate indicates faster data drift and lower reliability, the integrity score is negatively correlated with the trend change rate.

[0057] In the formula, This represents the integrity score; This is the preset sensitivity adjustment coefficient.

[0058] Next, the step of calculating the dynamic validity period of the UPD data is performed. This step is based on a preset reliability decay model, which includes a linear correction function and an exponential decay function. The design logic of this model is that the current quality (freshness) determines the basic length of the validity period, while the trend of change (completeness) determines the decay rate of the validity period. Substituting the freshness score into the linear correction function, the time correction coefficient is calculated. This process reflects the logic that the better the data quality, the longer the basic trust time.

[0059] In the formula, This represents the time correction factor; This is the preset linear gain coefficient.

[0060] Meanwhile, the integrity score is substituted into the exponential decay function to calculate the decay factor. This process reflects the logic that if the data shows a deterioration trend, its reliability duration will decrease exponentially.

[0061] In the formula, This represents the attenuation factor; This is the preset attenuation rate coefficient.

[0062] Finally, the product of the preset baseline validity period, the time correction coefficient, and the attenuation factor is calculated, and the calculated product is determined as the dynamic validity period of the UPD data.

[0063] In the formula, This indicates the dynamic validity period of the UPD data; This indicates the preset baseline validity period (e.g., the default aging time set based on experience).

[0064] Finally, the calculated freshness score, integrity score, and dynamic validity period of the UPD data are combined and output as a comprehensive reliability evaluation result to the subsequent strategy adjustment module.

[0065] The comprehensive reliability evaluation results generated through the above methods not only quantify the current instantaneous quality through freshness scoring and reveal potential evolution trends through integrity scoring, but also predict the future usability window of the data through dynamic validity period. This constructs a full-dimensional evaluation system that includes the past, present, and future, providing a scientific and quantitative decision-making basis for achieving refined positioning strategy adjustments.

[0066] Step S5: Compare the comprehensive reliability evaluation results with the preset benchmark, and perform a matching positioning strategy adjustment operation based on the comparison results.

[0067] In a preferred embodiment, comparing the comprehensive reliability evaluation result with a preset benchmark and performing a matching positioning strategy adjustment operation based on the comparison result includes: The freshness score is compared with a preset first quality threshold and a preset minimum usable threshold, respectively; the integrity score is compared with a preset second quality threshold; and the dynamic expiration date is compared with a preset time threshold. If the freshness score is between the minimum usable threshold and the first quality threshold, or the integrity score is lower than the second quality threshold, the operation of reducing the weight of the target satellite in the positioning solution model is performed. If the dynamic validity period is lower than the time threshold, execute the request and prefetch backup UPD data; If the freshness score is lower than the minimum available threshold, at least one of the following operations is performed: removing the target satellite and switching the data source; wherein the minimum available threshold is less than the first quality threshold.

[0068] Specifically, after obtaining the comprehensive reliability evaluation results of the UPD data, in order to achieve optimal closed-loop control of positioning performance, it is necessary to compare the comprehensive reliability evaluation results with a preset benchmark, determine the anomaly level based on the comparison results, and execute positioning strategy adjustment operations matching the anomaly level. The core logic of this step lies in establishing a mapping relationship from "multi-dimensional evaluation indicators" to "differentiated control decisions," avoiding the crude "black and white" processing of traditional methods. Instead, it adopts a graded response strategy based on subtle differences in data quality to achieve the best balance between the number of available satellites and positioning accuracy.

[0069] In a preferred embodiment, the comparison and decision-making process involves multi-dimensional threshold discrimination. The system presets a first quality threshold and a minimum availability threshold to measure real-time quality, wherein the minimum availability threshold is numerically strictly smaller than the first quality threshold, thereby constructing a buffer between "high-quality availability" and "complete unavailability" that allows for "degraded use." Simultaneously, the system also presets a second quality threshold to measure the evolving situation and a time threshold to measure the remaining effective duration. During execution, the processor compares the freshness score calculated in the preceding steps with the first quality threshold and the minimum availability threshold, respectively; simultaneously, it compares the integrity score with the second quality threshold and the dynamic validity period with the time threshold.

[0070] When the comparison results show that the freshness score is between the minimum usable threshold and the first quality threshold, or the integrity score is lower than the second quality threshold, it indicates that although the current UPD data is not completely invalid, its accuracy has shown signs of decline or a slowly deteriorating drift trend (i.e., it is in a sub-healthy state). At this time, the operation of reducing the weight of the target satellite in the positioning solution model is performed.

[0071] Specifically, in the Kalman filter-based positioning solution model, the weight reduction is achieved by adjusting the diagonal elements of the measurement noise covariance matrix (R matrix) corresponding to the target satellite. By artificially amplifying the variance of the satellite's observation noise, the filter gain is forced to decrease, thereby suppressing the magnitude of the correction to the final state estimate by the satellite's observation data.

[0072] The formula for calculating variance inflation is as follows: In the formula, This represents the adjusted measurement noise variance; This represents the original measurement noise variance (usually determined by the elevation angle model or the carrier-to-noise ratio model). This represents the preset variance inflation coefficient, and In this way, the target satellite's geometry (DOP value) is preserved while minimizing the negative impact of its anomalous data on the positioning results.

[0073] When the comparison result shows that the dynamic validity period is lower than the time threshold, it indicates that the currently used UPD data is about to expire. If it is not updated in time, the correction data will become invalid, causing location divergence. At this time, the operation of requesting and pre-fetching backup UPD data is performed. This operation is usually performed asynchronously in the background. That is, while maintaining the current location calculation, a new UPD data request is sent to the server, and the new data is cached in the backup queue. Once the current data reaches the end of its validity period or its quality deteriorates further, the pre-fetched backup UPD data can be used to achieve seamless switching, avoiding service interruption caused by network request latency.

[0074] When the comparison results show that the freshness score is lower than the minimum usable threshold, it indicates that the quality of the current UPD data has severely deteriorated, containing excessive bias or experiencing severe cycle slips. Continued use will directly disrupt positioning convergence. In this case, at least one of the following operations is performed: removing the target satellite or switching the data source. Removing the target satellite directly removes it from the current positioning observation equations and prevents it from participating in the solution. Switching the data source involves attempting to replace the current data stream with an alternative UPD product (e.g., UPD data streams from different analysis centers) and re-evaluating the stability of the replaced data. It should be noted that these three positioning strategy adjustment operations are not mutually exclusive and can be performed in parallel.

[0075] By adjusting the above-mentioned hierarchical strategy, the most appropriate countermeasures can be taken for data anomalies of different degrees, maximizing the accuracy and reliability of positioning calculation while ensuring the continuity of positioning services.

[0076] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0077] like Figure 2 As shown, an embodiment of the present invention provides a positioning strategy adjustment device based on UPD time series stability evaluation, including: a data acquisition module, a feature construction module, an index calculation module, a comprehensive evaluation module, and a strategy adjustment module; The data acquisition module is used to acquire multi-source time-series observation data of the target satellite during the real-time monitoring period; wherein, the multi-source time-series observation data includes UPD solution residual values; The feature construction module is used to construct a time-series feature vector sequence based on the multi-source time-series observation data; wherein, each sequence element in the time-series feature vector sequence is a feature vector of the corresponding epoch, and each dimension of the feature vector corresponds to a type of multi-source time-series observation data; The index calculation module is used to extract the feature component sequence corresponding to the UPD solution residual value in the time series feature vector sequence, calculate the first statistical dispersion of the feature component sequence in a preset short time window and the second statistical dispersion of the feature component sequence in a preset long time window; calculate the ratio of the first statistical dispersion to the second statistical dispersion, and determine the ratio as an instability index. The comprehensive evaluation module is used to combine the time-series feature vector sequence and the instability index to generate a comprehensive reliability evaluation result for UPD data; wherein, the comprehensive reliability evaluation result includes an availability score for UPD data in the current epoch and a prediction of the future effective duration of UPD data; The strategy adjustment module is used to compare the comprehensive reliability evaluation results with a preset benchmark, and perform a matching positioning strategy adjustment operation based on the comparison results.

[0078] It should be noted that the embodiments of the device described above correspond to the embodiments of the present invention described above, and can realize the positioning strategy adjustment method based on UPD timing stability evaluation described above in any one of the present invention. Furthermore, the embodiments of the device described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without creative effort.

[0079] Based on the above-described method embodiments of the present invention, a corresponding embodiment of an electronic device is provided.

[0080] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the positioning strategy adjustment method based on UPD timing stability evaluation as described in any one of the present invention, or the processor implements the functions of each module in the above-described device embodiments.

[0081] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0082] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0083] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0084] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0085] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments; Another embodiment of the present invention provides a storage medium including a stored computer program, wherein, when the computer program is running, it controls the device where the storage medium is located to execute any of the positioning strategy adjustment methods based on UPD timing stability evaluation described above.

[0086] The aforementioned storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0087] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0088] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A positioning strategy adjustment method based on UPD timing stability evaluation, characterized in that, The method comprises: acquiring multi-source time series observation data of a target satellite in a real-time monitoring period; wherein the multi-source time series observation data comprises UPD residual error values; constructing a time series feature vector sequence according to the multi-source time series observation data; wherein each sequence element in the time series feature vector sequence is a feature vector corresponding to an epoch, and each dimension component of the feature vector corresponds to one kind of multi-source time series observation data; extracting a feature component sequence corresponding to the UPD residual error values from the time series feature vector sequence, calculating a first statistical dispersion of the feature component sequence within a preset short time window and a second statistical dispersion of the feature component sequence within a preset long time window, respectively; calculating a ratio of the first statistical dispersion to the second statistical dispersion, and determining the ratio as an instability indicator; generating a comprehensive reliability evaluation result of UPD data by combining the time series feature vector sequence and the instability indicator; wherein the comprehensive reliability evaluation result comprises an availability score of UPD data at a current epoch and a prediction of a future effective time length of UPD data; comparing the comprehensive reliability evaluation result with a preset benchmark, and performing a positioning strategy adjustment operation according to the comparison result.

2. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 1, wherein, The multi-source time series observation data further comprises carrier-to-noise ratio values and multipath effect values; The method further comprises: extracting UPD residual error values, carrier-to-noise ratio values and multipath effect values corresponding to a current observation epoch for each observation epoch included in the multi-source time series observation data; performing normalization processing on the UPD residual error values, carrier-to-noise ratio values and multipath effect values corresponding to the current observation epoch, respectively, to obtain a normalized residual error component, a normalized carrier-to-noise ratio component and a normalized multipath component of the current observation epoch; splicing the normalized residual error component, the normalized carrier-to-noise ratio component and the normalized multipath component of the current observation epoch to generate a feature vector of the current observation epoch; arranging the generated feature vectors of multiple observation epochs in a sequence according to the order of observation time to construct a time series feature vector sequence.

3. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 2, wherein, The method further comprises: extracting dimension data corresponding to the normalized residual error component from each feature vector included in the time series feature vector sequence to form a single-dimensional residual feature sequence in the order of epochs; taking the current observation epoch as a cutoff time, extracting a first data segment corresponding to a preset short time window and a second data segment corresponding to a preset long time window from the residual feature sequence in a backward direction; calculating the standard deviation of all normalized residual error components in the first data segment, and determining the calculated standard deviation as the first statistical dispersion; Calculate the standard deviation of all normalized residual components in the second data segment, and determine the calculated standard deviation as the second statistical dispersion.

4. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 3, wherein, The comprehensive reliability evaluation result of the UPD data generated by combining the time sequence feature vector sequence and the instability index includes: A feature vector corresponding to the current observation epoch is extracted from the time sequence feature vector sequence, and the feature vector of the current observation epoch and the instability index are weighted and fused using a preset weighting coefficient to obtain a freshness score representing real-time quality of the data; Trend fitting is performed on the time sequence feature vector sequence, a trend change rate of the fitted curve is calculated, and an integrity score representing an evolution trend of the data is generated according to the trend change rate; Based on a preset reliability decay model, the freshness score and the integrity score are taken as input parameters to calculate a dynamic validity period of the UPD data; The freshness score, the integrity score, and the dynamic validity period of the UPD data are combined as a comprehensive reliability evaluation result.

5. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 4, wherein, The trend fitting of the time sequence feature vector sequence, the calculation of the trend change rate of the fitted curve, and the generation of the integrity score representing the evolution trend of the data include: Time sequence data corresponding to the normalized residual component is extracted from the time sequence feature vector sequence to construct a normalized residual component sequence; A least square method is used to perform polynomial fitting on the normalized residual component sequence to obtain a fitted curve function representing a trend of the residual change; A first derivative value of the fitted curve function at a time corresponding to the current observation epoch is calculated, and an absolute value of the first derivative value is determined as the trend change rate; The trend change rate is converted into the integrity score using a preset negative correlation mapping function; the larger the value of the trend change rate, the lower the integrity score.

6. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 5, wherein, The reliability decay model includes a linear correction function and an exponential decay function. Based on the preset reliability decay model, the freshness score and the integrity score are taken as input parameters to calculate a dynamic validity period of the UPD data, including: The freshness score is substituted into the linear correction function to calculate a time correction coefficient; The integrity score is substituted into the exponential decay function to calculate a decay factor; The product of a preset reference validity period, the time correction coefficient, and the decay factor is calculated, and the calculated product is determined as the dynamic validity period of the UPD data.

7. The positioning strategy adjustment method based on UPD timing stability evaluation of claim 6, wherein, The comprehensive reliability evaluation result is compared with a preset reference, and a matching positioning strategy adjustment operation is performed according to the comparison result, including: The freshness score is compared with a preset first quality threshold and a preset minimum available threshold, respectively; the integrity score is compared with a preset second quality threshold; and the dynamic validity period is compared with a preset time threshold; If the freshness score is between the minimum available threshold and the first quality threshold, or the integrity score is lower than the second quality threshold, an operation of reducing the weight of the target satellite in the positioning solution model is performed. if the dynamic validity period is lower than the time threshold, performing an operation of requesting and prefetching backup UPD data; if the freshness score is lower than the minimum usability threshold, performing at least one of an operation of discarding the target satellite and an operation of switching data sources; wherein the minimum usability threshold is less than the first quality threshold.

8. A positioning strategy adjustment device based on UPD timing stability evaluation, characterized in that, The method comprises: a data acquisition module, a feature construction module, an index calculation module, a comprehensive evaluation module, and a strategy adjustment module; The data acquisition module is configured to acquire multi-source time series observation data of a target satellite in a real-time monitoring period; wherein the multi-source time series observation data comprises UPD calculation residual values. The feature construction module is configured to construct a time series feature vector sequence according to the multi-source time series observation data; wherein each sequence element in the time series feature vector sequence is a feature vector corresponding to an epoch, and each dimension component of the feature vector corresponds to a kind of multi-source time series observation data. The index calculation module is configured to extract a feature component sequence corresponding to the UPD calculation residual values from the time series feature vector sequence, calculate a first statistical dispersion of the feature component sequence within a preset short time window and a second statistical dispersion of the feature component sequence within a preset long time window, respectively, calculate a ratio of the first statistical dispersion to the second statistical dispersion, and determine the ratio as an instability index. The comprehensive evaluation module is configured to generate a comprehensive reliability evaluation result of UPD data in combination with the time series feature vector sequence and the instability index; wherein the comprehensive reliability evaluation result comprises an availability score of UPD data at a current epoch and a prediction of a future effective duration of UPD data. The strategy adjustment module is configured to compare the comprehensive reliability evaluation result with a preset benchmark, and perform a matching positioning strategy adjustment operation according to the comparison result.

9. An electronic device, comprising: The storage medium comprises a stored computer program, wherein when the computer program runs, the device where the storage medium is located performs the positioning strategy adjustment method based on UPD time series stability evaluation according to any one of claims 1 to 7.

10. A storage medium, characterized by The storage medium comprises a stored computer program, wherein when the computer program runs, the device where the storage medium is located performs the positioning strategy adjustment method based on UPD time series stability evaluation according to any one of claims 1 to 7.

Images