Three-dimensional multi-source navigation information system based on Beidou

Abstract

The invention provides a Beidou-based three-dimensional multi-source navigation information system, and relates to the technical field of data processing, and the system comprises a Beidou satellite navigation positioning module which is used for providing basic positioning information including longitude, latitude, height and time data; the multi-sensor data acquisition module comprises a barometer, an accelerometer and a thermometer and is used for acquiring environmental data, including an air pressure value, acceleration change and temperature; and the data fusion processing module is used for dynamically updating and fusing the data of each sensor by adopting Bayesian maximum likelihood estimation and Kalman filtering, and smoothing and optimizing the positioning data through a forward and backward combined filtering algorithm and a smoothing algorithm to obtain a positioning result containing height and position information. The problems of low positioning precision and poor navigation effect in the urban canyon environment are effectively solved, and the positioning precision and the user experience in the complex environment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a three-dimensional multi-source navigation information system based on Beidou. Background Art

[0002] Traditional GNSS technologies, such as GPS or the early Beidou system, exhibit obvious limitations in urban canyon environments. Due to the height and spacing of buildings, satellite signals are easily blocked and affected by multipath effects, resulting in a decrease in positioning accuracy.

[0003] In densely built-up urban areas, when traditional GNSS positioning devices are between high-rise buildings on both sides of the street, the signals are often blocked, leading to unstable positioning accuracy and even positioning failures. Especially on narrow streets between high-rise buildings, GNSS receivers may not be able to receive enough satellite signals to meet the minimum number of satellites required for positioning.

[0004] In addition, wrong routes often occur during navigation. This is because the high-rise buildings and complex terrain cause severe interference to satellite signals, and devices that rely solely on satellite signals for navigation are difficult to accurately determine the user's position, thus generating incorrect navigation paths.

[0005] Single-sensor technologies, such as relying solely on an inertial navigation system (INS) or a barometer for positioning, also have obvious deficiencies. Although the inertial navigation system can navigate autonomously, its errors will accumulate over time, resulting in a decrease in long-term positioning accuracy. Although the barometer can provide vertical positioning information, its positioning accuracy is also affected by factors such as weather changes and temperature changes.

[0006] During long-term vehicle navigation, devices that rely solely on the inertial navigation system will gradually accumulate errors, especially in situations where the driving route is complex and changeable, and the error accumulation is more obvious. For example, on sections with continuous turns or uphill and downhill slopes, the positioning errors of the inertial navigation system may cause the vehicle to deviate from the predetermined route.

[0007] Under weather conditions with large barometric pressure changes, such as heavy rain or typhoon weather, the positioning accuracy of the barometer will be affected. This is because the barometer determines the altitude by measuring the atmospheric pressure, and the atmospheric pressure is greatly affected by weather changes. Therefore, under weather conditions with drastic barometric pressure changes, devices that rely solely on the barometer for vertical positioning may have large positioning errors. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a three-dimensional multi-source navigation information system based on Beidou, which improves the positioning accuracy in urban areas.

[0009] To solve the above technical problem, the technical solution of the present invention is as follows:

[0010] In the first aspect, a Beidou-based three-dimensional multi-source navigation information system includes:

[0011] A Beidou satellite navigation and positioning module for providing basic positioning information, including longitude, latitude, altitude, and time data;

[0012] A multi-sensor data acquisition module, including a barometer, an accelerometer, and a thermometer, for acquiring environmental data, including barometric pressure values, acceleration changes, and temperature;

[0013] A data fusion processing module for dynamically updating and fusing the data of each sensor by using Bayesian maximum likelihood estimation and Kalman filtering, and smoothing and optimizing the positioning data through forward and backward combined filtering algorithms and smoothing algorithms to obtain a positioning result containing altitude and position information;

[0014] A three-dimensional positioning module for realizing centimeter-level altitude positioning by using barometric pressure change information according to the temperature and barometric pressure ranges of the longitude and latitude areas of the current position of the Beidou satellite navigation and positioning module;

[0015] A three-dimensional navigation function module for realizing horizontal navigation and vertical navigation in scenarios of complex terrain, urban canyons, and urban viaducts according to the positioning result of altitude and position information;

[0016] An urban space stratification strategy module for stratifying the urban space according to barometric pressure data and floor information, and matching the barometric pressure data of the current position during navigation to determine the floor where it is located to realize the floor navigation function;

[0017] A user interaction interface module for displaying three-dimensional navigation information, including horizontal position, altitude, floor information, route planning, and navigation instructions.

[0018] Furthermore, by using Bayesian maximum likelihood estimation and Kalman filtering, dynamically updating and fusing the data of each sensor, and smoothing and optimizing the positioning data through forward and backward combined filtering algorithms and smoothing algorithms to obtain a positioning result containing altitude and position information, including:

[0019] According to the prior information and current observation data, use the Bayesian maximum likelihood estimation method to calculate the posterior probability distribution of the data of each sensor to obtain the final estimated value of the data of each sensor;

[0020] According to the final estimated values of the data of each sensor, establish a forward Kalman filtering model, and through iterative calculation, obtain the forward filtering state vector and error covariance matrix; according to the forward filtering state vector, rewrite the backward filtering model to obtain the backward filtering state vector and error covariance matrix;

[0021] At each moment, the state vector and error covariance of the combined filter are obtained through information fusion using the forward filtering state vector and the backward filtering state vector, and the smoothing algorithm is used to perform backward smoothing on the combined filtering result to obtain the smoothed state vector and covariance matrix; the initial value of the backward smoothing is selected as the state vector and covariance matrix at the end epoch of filtering.

[0022] The position information, including longitude, latitude, and altitude, is extracted from the smoothed state vector as the final positioning result.

[0023] Furthermore, the calculation process of the posterior probability distribution of each sensor data includes:

[0024] Based on Bayes' theorem, the posterior probability distribution is calculated through the joint probability of the sensor observation data and the prior distribution information, where:

[0025] For the observation data point of the i-th sensor, the numerator part of the exponential joint probability function is constructed as a weighted combination of the observation data deviation term and the parameter deviation term. The observation data deviation term is the normalized squared difference of the variance between the current measurement value and the parameter-related expected mean, and the parameter deviation term is the normalized squared difference of the prior variance between the current parameter value and the prior mean.

[0026] The denominator part of the posterior probability distribution is constructed as a product combination of the normalization integral factor and the observation variance and prior variance constant terms, where the normalization integral factor realizes probability normalization through the global integration of the parameter space.

[0027] The variance calculation of the observation data adopts the finite sample unbiased estimation method. Specifically, the sample mean of the n measurement values of the i-th sensor is calculated, and then the sum of the squared deviations of each measurement value from the sample mean is divided by the degree-of-freedom parameter. The parameter prior mean and prior variance are determined according to the sensor historical calibration data, where the prior mean represents the expected central value of the parameter, and the prior variance represents the uncertainty range of the parameter estimation. The probability distribution calculation process finally outputs the parameter posterior probability distribution of each sensor under the given observation data conditions for the iterative optimization of the subsequent Bayesian maximum likelihood estimation.

[0028] Furthermore, according to the temperature and pressure ranges in the longitude and latitude regions of the current position of the Beidou satellite navigation positioning module, centimeter-level altitude positioning is achieved using barometric change information, including:

[0029] Decode the positioning signal of the Beidou satellite to obtain the longitude and latitude coordinates, speed, and time information of the current position, and obtain the temperature and pressure ranges in the longitude and latitude regions of the current position from the meteorological data source.

[0030] Establish a standard atmosphere model based on the relationship between air pressure and altitude, and adjust the parameters of the standard atmosphere model according to the changes in temperature and air pressure;

[0031] Calculate a preliminary altitude value based on the air pressure data at the current location;

[0032] Take the known altitude point as the reference location, calculate the air pressure change rate between the current location and the reference location, and use the standard atmosphere model to correct the preliminary altitude value according to the air pressure change rate to obtain the corrected altitude value;

[0033] Measure the air pressure at the current location using an air pressure sensor and combine it with the corrected altitude value to achieve centimeter-level altitude positioning.

[0034] Furthermore, based on the positioning results of altitude and location information, implement horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios, including:

[0035] Receive Beidou signals, obtain the altitude and location information of the current location, identify the terrain type where the user is located using map data and real-time sensor information, and at the same time combine Beidou signals to achieve positioning in three-dimensional space, including horizontal position and vertical altitude;

[0036] According to the destination input by the user and the current location, combine map data and real-time traffic information to plan the final horizontal navigation path;

[0037] During navigation, provide real-time navigation guidance to the user in visual and voice ways, including turning prompts, road names, distance information, and dynamically adjust the navigation path according to real-time traffic conditions and road conditions changes;

[0038] After navigation, according to the user's needs and the altitude information of the current location, plan the altitude change range of vertical navigation, and during the planning process, combine Beidou signals and the data of altitude sensors to achieve horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios.

[0039] Furthermore, according to the air pressure data and floor information, stratify the urban space, and match the air pressure data at the current location during navigation to determine the floor where the user is located to achieve the floor navigation function, including:

[0040] Based on the air pressure data inside the building, establish a mapping relationship between air pressure and floors by measuring air pressure at different floors and recording the corresponding floor information to obtain an air pressure - floor mapping table;

[0041] At the start of navigation, obtain the air pressure data at the current location and match the current air pressure data with the air pressure - floor mapping table to determine the initial floor where the user is located;

[0042] Plan a navigation path from the current floor to the destination floor according to the destination floor input by the user and the initially matched current floor;

[0043] During the execution of navigation, continuously monitor the barometric pressure data at the current location, and update the floor information in real time according to the barometric pressure change. If the current floor information is different from the matched result, re-plan the navigation path according to the new floor information;

[0044] Provide real-time floor navigation guidance to the user in visual and voice ways, including the current floor name, the distance information from the destination floor, and turning prompts. The user moves between floors according to the floor navigation guidance;

[0045] When the user arrives at the destination floor, issue an arrival prompt and end the navigation process. If an abnormal situation is encountered during the navigation process, handle it according to the preset abnormal handling mechanism, including prompting the user to manually input the floor information to continue navigation, or switching to the floor navigation method according to the map.

[0046] In a second aspect, a computing device includes:

[0047] One or more processors;

[0048] A storage device for storing one or more programs, which when executed by the one or more processors cause the one or more processors to implement the system described above.

[0049] In a third aspect, a computer-readable storage medium stores a program that implements the system when executed by a processor.

[0050] The above solution of the present invention has at least the following beneficial effects:

[0051] Integrates Beidou satellite navigation and positioning data and data from various sensors such as barometers, accelerometers, and thermometers. Through Bayesian maximum likelihood estimation and Kalman filtering techniques, dynamically update and fuse the data of each sensor, effectively overcoming the limitations of a single sensor in a complex environment. The three-dimensional positioning module uses barometric pressure change information to achieve centimeter-level height positioning, significantly improving the positioning accuracy in the vertical direction. The data fusion processing module uses forward and backward combined filtering algorithms and smoothing algorithms to smooth and optimize the positioning data, further improving the overall positioning accuracy and stability.

[0052] Based on the positioning results of height and position information, the three-dimensional navigation function module realizes horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios. This function has great practical value for users who need to navigate in areas with dense high-rise buildings or cities with complex terrain. The urban space stratification strategy module stratifies the urban space according to air pressure data and floor information, and matches the air pressure data of the current position during navigation to determine the floor where it is located, realizing the floor navigation function, which further improves the accuracy and convenience of navigation.

[0053] The user interface module displays three-dimensional navigation information in an intuitive and easy-to-use manner, including horizontal position, height, floor information, route planning, and navigation instructions, etc., enabling users to easily understand their current position and navigation path. The three-dimensional navigation function of the system provides users with a more comprehensive and accurate navigation service, especially in scenarios that require vertical navigation (such as inside high-rise buildings or urban viaducts), greatly enhancing the user's navigation experience.

[0054] This system is not only applicable to the navigation needs of individual users, but also can be widely used in multiple fields such as intelligent transportation, unmanned driving, logistics distribution, and mobile services. For example, in an intelligent transportation system, this system can provide high-precision vehicle positioning information, optimize driving routes, and improve traffic efficiency; in unmanned driving technology, this system can provide reliable positioning support for autonomous vehicles, improving safety and stability. The research and development and application of this system have promoted the development of multi-sensor fusion technology, data fusion algorithms, and three-dimensional navigation and positioning technology. By continuously optimizing and improving these technologies, it can provide strong support for more complex and diverse future navigation and positioning needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a schematic diagram of a Beidou-based three-dimensional multi-source navigation information system provided by an embodiment of the present invention.

[0056] Figure 2 is a schematic flow diagram of realizing horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios based on the positioning results of height and position information of the Beidou-based three-dimensional multi-source navigation information system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0058] As shown Figure 1 in the figure, an embodiment of the present invention provides a Beidou-based three-dimensional multi-source navigation information system, including:

[0059] A Beidou satellite navigation and positioning module 11, which is used to provide basic positioning information, including longitude, latitude, altitude and time data;

[0060] A multi-sensor data acquisition module 12, including a barometer, an accelerometer and a thermometer, which is used to acquire environmental data, including air pressure value, acceleration change and temperature;

[0061] A data fusion processing module 13, which is used to adopt Bayesian maximum likelihood estimation and Kalman filtering to dynamically update and fuse the data of each sensor, and through forward and backward combined filtering algorithms and smoothing algorithms, smooth and optimize the positioning data to obtain a positioning result including altitude and position information;

[0062] A three-dimensional positioning module 14, which is used to utilize the air pressure change information to achieve centimeter-level altitude positioning according to the temperature and air pressure ranges of the longitude and latitude areas of the current position of the Beidou satellite navigation and positioning module;

[0063] A three-dimensional navigation function module 15, which is used to achieve horizontal navigation and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios according to the positioning results of altitude and position information;

[0064] An urban space stratification strategy module 16, which is used to stratify the urban space according to air pressure data and floor information, and match the air pressure data of the current position during navigation to determine the floor where it is located, so as to achieve the floor navigation function;

[0065] A user interaction interface module 17, which is used to display three-dimensional navigation information, including horizontal position, altitude, floor information, route planning and navigation instructions.

[0066] In the embodiment of the present invention, the Beidou satellite navigation and positioning module 11 provides high-precision longitude, latitude, altitude and time data as the basic positioning information of the system. The high precision and anti-interference ability of the Beidou system ensure the stability and reliability of the positioning data.

[0067] The multi-sensor data acquisition module 12 acquires environmental data, such as air pressure value, acceleration change and temperature, through a barometer, an accelerometer and a thermometer, providing the system with rich environmental perception ability. These data complement the Beidou satellite positioning data, enhancing the robustness of the positioning system.

[0068] The data fusion processing module 13 adopts Bayesian maximum likelihood estimation and Kalman filtering techniques to dynamically update and fuse the data of each sensor. Through the forward and backward combined filtering algorithm and the smoothing algorithm, the positioning data is further smoothed and optimized, effectively reducing the impact of the data fluctuation of a single sensor on the overall positioning result, and improving the positioning accuracy and stability.

[0069] The three-dimensional positioning module 14 combines the data of the Beidou satellite navigation positioning module and the barometric pressure change information collected by the barometer to achieve centimeter-level altitude positioning, providing the system with high-precision three-dimensional positioning capabilities.

[0070] The three-dimensional navigation function module 15 realizes horizontal navigation and vertical navigation in scenarios such as complex terrains, urban canyons, and urban viaducts according to the positioning results of altitude and position information. This function facilitates the travel needs of users in complex environments.

[0071] The urban space stratification strategy module 16 uses barometric pressure data and floor information to stratify the urban space, and matches the barometric pressure data at the current position during navigation to determine the floor where it is located, realizing the floor navigation function. This function is particularly useful in areas with dense high-rise buildings in the city, providing more accurate navigation services for users.

[0072] The user interaction interface module 17 displays three-dimensional navigation information in an intuitive and easy-to-use manner, including horizontal position, altitude, floor information, route planning, and navigation instructions, etc. Users can understand the current position and navigation path in real time through this interface, greatly enhancing the user's navigation experience and interactivity.

[0073] In a preferred embodiment of the present invention, the Beidou satellite navigation positioning module 11 is used to provide basic positioning information, including longitude, latitude, altitude, and time data; the multi-sensor data acquisition module 12, including a barometer, an accelerometer, and a thermometer, is used to acquire environmental data, including barometric pressure values, acceleration changes, and temperatures, and may include:

[0074] In the embodiment of the present invention, the Beidou satellite navigation positioning module is started for initialization settings, including configuring communication interfaces (such as UART, SPI, etc.), setting data output formats (such as NMEA protocol format), setting positioning frequencies, etc., to ensure that the module can normally receive Beidou satellite signals, perform self-checks and status monitoring, and ensure that the module is in a normal working state.

[0075] The Beidou satellite navigation and positioning module receives signals from Beidou satellites through an antenna, including navigation messages, pseudorange, carrier phase and other information. Decode and demodulate the received signals, and extract useful positioning information such as longitude, latitude, altitude and time data. Use the extracted positioning information, combined with the algorithms and models of the Beidou system, to perform positioning calculations. This includes steps such as calculating the satellite position, calculating the distance from the satellite to the receiver, and resolving the receiver position. Through multiple measurements and calculations, improve the positioning accuracy and reduce errors. The Beidou system uses technologies such as differential positioning and carrier phase measurement to improve the positioning accuracy.

[0076] Output the calculated positioning information (longitude, latitude, altitude and time data) in a predetermined format for use by other modules. The output format can be the standard NMEA protocol format or other custom formats. Regularly monitor and maintain the Beidou satellite navigation and positioning module, check its working status and performance indicators, and if any faults or performance degradation are found in the module, repair or replace it in a timely manner.

[0077] Select suitable barometers, accelerometers and thermometers to ensure that their accuracy, stability and reliability meet the system requirements. Integrate these sensors into the multi-sensor data acquisition module, and perform hardware connection and circuit debugging. Perform initialization settings for each sensor, including configuring the communication interface, setting the sampling frequency, calibrating the zero point, etc. The barometer collects the air pressure value of the current environment, the accelerometer collects the acceleration change of the object, and the thermometer collects the temperature of the current environment. These sensors collect data at a certain sampling frequency and transfer the data to the data fusion processing module for processing.

[0078] In a preferred embodiment of the present invention, Bayesian maximum likelihood estimation and Kalman filtering are used to dynamically update and fuse the data of each sensor, and through the forward-backward combined filtering algorithm and the smoothing algorithm, smooth and optimize the positioning data to obtain a positioning result including altitude and position information, which may include:

[0079] The calculation process of the posterior probability distribution of the data of each sensor includes:

[0080] Based on Bayes' theorem, calculate the posterior probability distribution through the joint probability of the sensor observation data and the prior distribution information, where:

[0081] For the observation data point of the i-th sensor, construct the numerator part of the exponential joint probability function as a weighted combination of the observation data deviation term and the parameter deviation term. The observation data deviation term is the normalized squared difference of the variance between the current measurement value and the parameter-related expected mean, and the parameter deviation term is the normalized squared difference of the prior variance between the current parameter value and the prior mean;

[0082] The denominator part of the posterior probability distribution is constructed as a product combination of a normalization integral factor, the observation variance, and the prior variance constant term, where the normalization integral factor realizes probability normalization through the global integration of the parameter space;

[0083] The variance of the observed data is calculated using the finite-sample unbiased estimation method. Specifically, the sample mean of the n measurement values of the i-th sensor is calculated, and then the sum of the squared deviations of each measurement value from the sample mean is divided by the degrees of freedom parameter. The prior mean and prior variance of the parameter are determined based on the historical calibration data of the sensor, where the prior mean represents the expected central value of the parameter, and the prior variance represents the uncertainty range of the parameter estimation. The probability distribution calculation process finally outputs the posterior probability distribution of the parameters of each sensor under the given observed data, which is used for the iterative optimization of the subsequent Bayesian maximum likelihood estimation.

[0084] When specifically applied, it can be implemented through the following specific calculation example, as follows:

[0085] According to the prior information and the current observed data, use the Bayesian maximum likelihood estimation method to calculate the posterior probability distribution of each sensor data, and obtain the final estimated value of each sensor data. The calculation formula for the posterior probability distribution of each sensor data is:

[0086]

[0087] where, P(Θ i ∣X i ) represents the probability distribution of the observed data X i of the i-th sensor when the parameter Θ i is known; X i represents the i-th observed data point; μ i (Θ i ) represents the expected mean of the observed data when the given parameter Θ i is known; Θ i represents the i-th parameter; represents the variance of the observed data, where, n represents the number of measurement values; j represents the index variable; i represents the index of the sensor; X ij represents the j-th measurement value of the i-th sensor; represents the mean of all measurement values of the i-th sensor; represents the prior mean of the parameter Θ i ; represents the prior variance; dΘ i represents the infinitesimal change amount when integrating the parameter Θ i ; π represents the constant pi;

[0088] Based on the final estimated values of the sensor data, a forward Kalman filter model is established. Through iterative calculations, the forward filtering state vector is obtained. and the error covariance matrix P f,k ; According to the forward filtering state vector, the backward filtering model is rewritten to obtain the backward filtering state vector and the error covariance matrix P r,k ; The backward filtering model is: Where represents the inverse of the state transition matrix in the backward filtering; X k+1 represents the state vector at time k + 1; represents the inverse of the noise distribution matrix in the backward filtering; W k represents the process noise vector at time k; represents the inverse of the state transition matrix from time k to time k + 1; γ k represents the noise distribution matrix at time k; k represents the current epoch; X k represents the state vector at time k; φ k / (k-1) represents the state transition from time k - 1 to time k; X k-1 represents the state vector at time k - 1; γ k-1 represents the noise distribution matrix; W k-1 represents the noise vector; Z k represents the observed value at time k; H k represents the design matrix; V k represents the observation noise vector;

[0089] At each time k, using the forward filtering state vector and the backward filtering state vector The state vector and error covariance of the combined filtering are obtained through information fusion And the reverse smoothing is performed on the combined filtering result using the smoothing algorithm to obtain the smoothed state vector and covariance matrix

[0090]

[0091] Where P f,k represents the error covariance matrix of the forward filtering at time k; represents the smoothed state vector; P s,k represents the smoothed covariance matrix; K s,k represents the gain matrix; represents the transpose of the forward filtering state transition matrix; represents the inverse of the error covariance matrix from time k to time k + 1 in the forward filtering; represents the state vector at the next moment k+1 after smoothing; represents the state vector predicted from moment k to moment k+1 in the forward filtering; K s,k represents the gain matrix; P s,(k+1) represents the error covariance matrix at the next moment k+1 after smoothing; P f,(k+1) / k represents the error covariance matrix from moment k to moment k+1 in the forward filtering; represents the transpose of the gain matrix; represents the error covariance matrix obtained by the combined filtering; represents the inverse of the error covariance matrix of the forward filtering; represents the inverse of the error covariance matrix of the backward filtering; represents the state vector obtained by the combined filtering; P c,k represents the error covariance matrix obtained by the combined filtering;

[0092] The initial value of the backward smoothing is selected as the state vector and covariance matrix at the end epoch of the filtering, that is and P s,e = P f,e ; where represents the state vector obtained by the smoothing algorithm at the end epoch of the filtering; represents the state vector obtained by the forward filtering at the end epoch of the filtering; P s,e represents the error covariance matrix obtained by the smoothing algorithm at the end epoch of the filtering; P f,e represents the error covariance matrix obtained by the forward filtering at the end epoch of the filtering;

[0093] Extract the position information, including longitude, latitude, and altitude, from the smoothed state vector as the final positioning result.

[0094] In the embodiments of the present invention, variables and parameters are defined: X i is the observation data of the i-th sensor, Θ i is the parameter of the i-th sensor, μ i (Θ i ) is the expected mean of the observation data when the given parameter is Θ i , is the variance of the observation data, and the calculation formula is where n is the number of measurement values, X ij is the j-th measurement value of the i-th sensor, is the mean of all measurement values of the i-th sensor. is the prior mean of the parameter Θ i ; is the prior variance of the parameter Θ i ;

[0095] Calculate the likelihood of the observed data Calculate the prior probability of the parameters Combine the likelihood and the prior to obtain the numerator of the posterior probability Calculate the normalization constant (denominator), including the integral over Θ i of Obtain the parameter Θ by maximizing the posterior probability distribution i estimate value of

[0096] Initialize the state vector and the error covariance matrix P f,0 . Set according to the prior knowledge of the system. State prediction where is the prediction of the system state at time step k, based on the state estimate at time step k - 1; is the state estimate at time step k - 1.

[0097] Error covariance prediction where Q k-1 is the process noise covariance matrix. Update the Kalman gain where R k is the measurement noise covariance matrix; P f,l∣l-1 is the covariance matrix of the state prediction error at time step k; is the measurement matrix H k transpose matrix of; H k is the measurement matrix. State update where is the updated estimate of the system state at time step k; K k is the Kalman gain. Error covariance update P f,k =(I - K f,k H k )P f,k∣k-1 ; where P f,k is the error covariance matrix of the state estimate at time step k; I is the identity matrix, with elements on the diagonal being 1 and other elements being 0.

[0098] Initialize the state vector X of the backward filtering r,N and the error covariance matrix P r,N as the final result of the forward filtering. State prediction where is the predicted state vector at time k, based on the backward filtering; is the state estimate value of the backward filtering at time k + 1. Error covariance prediction P r,k∣k+1is the error covariance matrix predicted based on backward filtering at time k; P r,k+1 is the error covariance matrix of backward filtering at time k+1; is the inverse mapping of the state transition matrix; Q k is the covariance matrix of the system process noise; is the noise input matrix in the inverse mapping.

[0099] Backward Kalman gain update where K r,k is the backward Kalman gain matrix; P r,k∣k+1 is the error covariance matrix of backward prediction. State update where, is the state estimate of backward filtering; is the prediction of the state at time k based on the information at time k+1; K r,k is the backward Kalman gain matrix. Error covariance update P r,k =(I-K r,k H k )P r,k∣k+1 . State vector and error covariance matrix and where, is the smoothed error covariance matrix; is the inverse of the forward filtering error covariance matrix; is the inverse of the backward filtering error covariance matrix; is the smoothed state vector; is the state estimate of forward filtering; is the state estimate of backward filtering. Use the state vector and covariance matrix at the end epoch of filtering as the initial values for smoothing, i.e., and P s,e =P f,e .

[0100] Smoothing step:

[0101] Gain matrix

[0102] State smoothing

[0103] Error covariance smoothing

[0104] Extract longitude, latitude, and altitude information from the smoothed state vector as the final positioning result.

[0105] Assume the following sensor data is collected at a certain moment:

[0106] Barometer reading: 990 hPa, accelerations on the X, Y, and Z axes are 0.1 m / s 2 , 0.2 m / s 2 , -9.8 m / s 2 (considering the acceleration due to gravity), thermometer reading 25 °C. Meanwhile, assume that the Beidou satellite navigation and positioning module provides the following prior information:

[0107] Longitude: 116.39°E;

[0108] Latitude: 39.90°N;

[0109] Altitude: 50 m;

[0110] Time: the current moment;

[0111] Through Bayesian maximum likelihood estimation, the posterior probability distribution of each sensor data can be calculated, and their final estimated values can be obtained. Then, using these estimated values to establish a forward Kalman filter model, the forward filtering state vector and error covariance matrix can be obtained through iterative calculation. According to the forward filtering results, the backward filtering model is rewritten to obtain the backward filtering state vector and error covariance matrix. At each moment, using the forward and backward filtering results, the state vector and error covariance of the combined filtering are obtained through information fusion. Finally, the smoothing algorithm is used to perform backward smoothing on the combined filtering results to obtain the smoothed state vector. From the smoothed state vector, the longitude, latitude, and altitude information can be extracted as the final positioning result.

[0112] Through Bayesian maximum likelihood estimation and Kalman filtering technology, the system can dynamically update and fuse the data of each sensor, effectively reducing the impact of single-sensor data fluctuations and errors on the positioning result, thereby improving the positioning accuracy. The Bayesian maximum likelihood estimation method can utilize prior information and current observation data to calculate the posterior probability distribution of each sensor data, improving the system's robustness to noise and outliers. Kalman filtering technology can achieve optimal estimation under the linear Gaussian assumption, adapt to dynamic environments, and improve the system's stability and reliability. Through the forward and backward combined filtering algorithm and the smoothing algorithm, the system can smooth and optimize the positioning data, achieving high-precision three-dimensional positioning, including longitude, latitude, and altitude information. The high-precision positioning result and three-dimensional navigation function can provide users with more accurate and intuitive navigation services, enhancing the user experience and satisfaction.

[0113] In a preferred embodiment of the present invention, according to the temperature and pressure ranges of the longitude and latitude regions of the current position of the Beidou satellite navigation and positioning module, using the barometric change information to achieve centimeter-level altitude positioning may include:

[0114] Decode the positioning signals of Beidou satellites to obtain the longitude and latitude coordinates, speed, and time information of the current location, and obtain the temperature and pressure ranges of the longitude and latitude area of the current location from meteorological data sources;

[0115] Establish a standard atmosphere model based on the relationship between pressure and altitude, and adjust the parameters of the standard atmosphere model according to the changes in temperature and pressure;

[0116] Use the barometric height formula And calculate the preliminary height value based on the pressure data of the current location; where h represents height; R represents the gas constant; U represents temperature; g represents the acceleration due to gravity; P0 represents the sea-level pressure; P represents the current pressure;

[0117] Take the known height point as the reference position and calculate the pressure change rate between the current position and the reference position And correct the preliminary height value using the standard atmosphere model according to the pressure change rate to obtain the corrected height value; where P represents the current pressure; P b represents the pressure value of the reference position; Δh represents the height difference between the current position and the reference position;

[0118] Measure the pressure of the current location using a barometric pressure sensor and combine it with the corrected height value to achieve centimeter-level height positioning.

[0119] In the embodiments of the present invention, ensure that the Beidou satellite navigation module is normally connected and working, support receiving and processing Beidou satellite signals, receive Beidou satellite signals through the antenna, and transmit the signals to the navigation module. The software in the navigation module decodes the received signals, extracts information such as longitude and latitude coordinates, speed, and time, and outputs the decoded information in a data format for subsequent processing. Determine a suitable meteorological data source, such as a weather station, an online weather service API, etc. Send a request to the meteorological data source to obtain the temperature and pressure ranges corresponding to the longitude and latitude of the current location, parse the data returned by the meteorological data source, and extract the temperature and pressure values. Select a suitable standard atmosphere model, such as the International Standard Atmosphere (ISA) model. Adjust the parameters of the standard atmosphere model, such as the temperature gradient, pressure change rate, etc., according to the temperature and pressure values of the current location.

[0120] Use the barometric height formula where R is the gas constant, U is the temperature, g is the acceleration due to gravity, P0 is the sea-level pressure, and P is the current pressure. Substitute the obtained temperature and pressure values into the formula and perform the calculation to obtain the preliminary height value. Select a reference position with a known height and calculate the pressure change rate between the current position and the reference position where P is the current pressure, P bThe air pressure value at the reference position, and Δh is the height difference between the current position and the reference position. According to the air pressure change rate, the preliminary height value is corrected using the standard atmosphere model to obtain the corrected height value. The air pressure at the current position is measured using a barometric pressure sensor, and combined with the corrected height value and the measurement value of the barometric pressure sensor, centimeter-level height positioning is achieved.

[0121] Suppose a Beidou satellite navigation and positioning module is installed on an unmanned aerial vehicle (UAV) to achieve centimeter-level height positioning. The Beidou module on the UAV receives Beidou satellite signals and decodes the longitude and latitude coordinates of the current position as (39.9042, 116.4074), the speed is 5 m / s, and the time is 12:00:00. The current temperature at the current position is obtained through the online weather service API as 25 °C, and the air pressure range is 1000 - 1010 hPa. The international standard atmosphere model is selected, and the model parameters are adjusted according to the temperature and air pressure values. The preliminary height value is calculated using the barometric height formula as 100 m. A reference position with a known height is selected (such as the ground height is 0 m and the air pressure is 1010 hPa), the air pressure change rate is calculated as -0.1 hPa / m, and the preliminary height value is corrected according to the standard atmosphere model to obtain the corrected height value of 99.5 m. Combining with the measurement value of the barometric pressure sensor (1002 hPa), centimeter-level height positioning is achieved, and finally the UAV height is determined to be 99.55 m.

[0122] By combining air pressure change information and the standard atmosphere model, centimeter-level height positioning is achieved, improving the positioning accuracy. Using a barometric pressure sensor to measure the current air pressure and combining with the corrected height value enhances the stability and reliability of the system. It can adapt to the height positioning requirements under different temperature and air pressure conditions, improving the adaptability and flexibility of the system. It provides more accurate height positioning services for fields such as UAVs, autonomous driving vehicles, and aerospace, promoting the development and innovation of related applications.

[0123] In a preferred embodiment of the present invention, according to the positioning results of height and position information, horizontal navigation and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios can be achieved, which may include:

[0124] Receive Beidou signals, obtain the height and position information of the current position, and use map data and real-time sensor information to identify the type of terrain where it is currently located. At the same time, combined with Beidou signals, positioning in three-dimensional space is achieved, including horizontal position and vertical height;

[0125] According to the destination input by the user and the current position, combined with map data and real-time traffic information, plan the final horizontal navigation path;

[0126] During the navigation process, provide real-time navigation guidance to the user through visual and voice means, including turning prompts, road names, and distance information, and dynamically adjust the navigation path according to real-time traffic conditions and road conditions changes;

[0127] After navigation, according to the user's needs and the current position altitude information, plan the altitude change range of vertical navigation, and during the planning process, combine the data of Beidou signals and altitude sensors to achieve horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios.

[0128] In the embodiment of the present invention, receive Beidou satellite signals through an antenna, decode the signals using software or firmware in the navigation module, and extract the altitude and longitude and latitude coordinates of the current position. Integrate the decoded altitude and position information with map data and real-time sensor information (such as accelerometers, gyroscopes, etc.) to prepare for subsequent terrain recognition and three-dimensional positioning. Load the terrain data around the current position from the pre-stored map database, and combine real-time sensor information (such as altitude change rate, acceleration, etc.) to make a preliminary judgment on the terrain. Use terrain recognition algorithms, such as classification algorithms based on elevation maps, to accurately identify the type of terrain where the user is currently located, such as complex terrain, urban canyons, urban viaducts, etc.

[0129] Utilize the altitude information and longitude and latitude coordinates provided by Beidou signals, combined with the terrain recognition results, to calculate the accurate coordinates of the current position in three-dimensional space.

[0130] Assume that Lat is the latitude coordinate provided by Beidou signals (unit: degree); Lon is the longitude coordinate provided by Beidou signals (unit: degree); AltGPS is the altitude information provided by Beidou signals (unit: meter), relative to the ellipsoid; AltTerrain is the altitude information obtained from terrain recognition (unit: meter), relative to the ellipsoid or sea level (specifically depending on the reference of the DEM).

[0131] Then the accurate coordinates of the current position in three-dimensional space are:

[0132] X, Y, Z coordinates (in the Earth-fixed coordinate system ECEF):

[0133] First, convert the longitude and latitude coordinates to radians:

[0134]

[0135] Among them, is the latitude (unit: radian), and λ is the longitude (unit: radian).

[0136] Then, use the parameters of the ellipsoid (such as the semi-major axis a and flattening f of the WGS-84 ellipsoid) to calculate the first eccentricity e of the ellipsoid 2= 2f - f 2 ; Next, calculate the radius of curvature of the ellipsoid

[0137] Finally, calculate the ECEF coordinates:

[0138]

[0139] If the height provided by the terrain database is relative to sea level, and the height provided by the Beidou signal is relative to the ellipsoid surface, the difference between the two needs to be considered during the calculation. The coordinates calculated by the above formula are in the ECEF coordinate system. If conversion to other coordinate systems (such as the ENU local coordinate system) is required, further coordinate transformation is also needed.

[0140] According to real-time sensor information and map data, continuously update and calibrate the three-dimensional coordinates to ensure the accuracy of positioning. Receive and parse the destination information input by the user, apply path planning algorithms (such as the Dijkstra algorithm), combine map data and real-time traffic information (such as road conditions, congestion situations, construction information, etc.) to plan the final horizontal navigation path from the current location to the destination. Dynamically optimize and adjust the planned path according to real-time traffic conditions and road condition changes. Design an intuitive and easy-to-use navigation interface to display information such as the current location, destination, and path planning. Provide real-time navigation guidance to the user through visual (such as arrow indications, road name displays, etc.) and voice (such as steering prompts, distance information, etc.) methods. Receive the user's interactive feedback (such as confirmation, cancellation, replanning, etc.), and adjust the navigation strategy according to the feedback. Analyze the user's needs for vertical navigation (such as parking in the garage, floor switching, etc.). According to the user's needs and the current location height information, combine map data and real-time sensor information to plan the height change range of vertical navigation.

[0141] During the planning process, combine the data of Beidou signals and height sensors to achieve horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios. Integrate the height information and longitude and latitude coordinates provided by Beidou signals with the data of height sensors to improve the accuracy and stability of navigation. Adjust the navigation strategy and path planning algorithm according to the characteristics of different scenarios such as complex terrain, urban canyons, and urban viaducts to ensure smooth and safe navigation.

[0142] Suppose an autonomous vehicle is equipped with a Beidou satellite navigation module and an altitude sensor, and is driving on an urban viaduct, planning to go to a destination on the ground floor. The Beidou module on the vehicle receives Beidou satellite signals and decodes the altitude of the current position as 50m, and the longitude and latitude coordinates as (121.4737, 31.2304). Combining map data and real-time sensor information, the vehicle identifies that it is currently in an urban viaduct scenario. Using the altitude information and longitude and latitude coordinates provided by the Beidou signal, combined with the terrain recognition result, the accurate coordinates of the vehicle in the three-dimensional space are calculated. According to the destination input by the user and the current position, combined with map data and real-time traffic information, the vehicle plans the optimal horizontal navigation path from under the viaduct to the destination. During the navigation process, the vehicle provides real-time navigation guidance to the driver through visual and voice means, including steering prompts, road names, distance information, etc. After reaching the exit of the viaduct, according to the driver's needs and the current position altitude information, the vehicle plans the altitude change range for vertical navigation (from 50m to the ground floor). Within the planned altitude change range, the vehicle combines the Beidou signal and the data of the altitude sensor to achieve horizontal navigation and vertical navigation in complex terrain and urban viaduct scenarios, and finally arrives at the destination safely.

[0143] By combining Beidou signals, map data and real-time sensor information, precise positioning and navigation in three-dimensional space are achieved, improving the accuracy of navigation. For different terrains and scenarios (such as complex terrain, urban canyons, urban viaducts, etc.), the navigation strategy and path planning algorithm can be automatically adjusted, enhancing the adaptability to scenarios. Real-time navigation guidance is provided to users through visual and voice means, improving the user experience and convenience. It provides more accurate and reliable navigation services for autonomous vehicles, promoting the development and application of autonomous driving technology.

[0144] In a preferred embodiment of the present invention, according to the air pressure data and floor information, the urban space is stratified, and the air pressure data of the current position is matched during navigation to determine the floor where it is located, and the floor navigation function can include:

[0145] According to the air pressure data inside the building, by measuring the air pressure on different floors and recording the corresponding floor information, a mapping relationship between air pressure and floors is established to obtain an air pressure - floor mapping table;

[0146] At the start of navigation, obtain the air pressure data of the current position, and match the current air pressure data with the air pressure - floor mapping table to determine the initial floor where the user is located;

[0147] According to the destination floor input by the user and the currently matched initial floor, plan the navigation path from the current floor to the target floor;

[0148] During the navigation process, continuously monitor the barometric pressure data at the current location, and update the floor information in real time according to the barometric pressure changes. If the current floor information is different from the matching result, re-plan the navigation path according to the new floor information;

[0149] Provide real-time floor navigation guidance to the user through visual and voice means, including the current floor name, the distance information to the target floor, and turning prompts. The user moves between floors according to the floor navigation guidance;

[0150] When the user reaches the target floor, issue an arrival prompt and end the navigation process. If an abnormal situation is encountered during the navigation process, handle it according to the preset abnormal handling mechanism, including prompting the user to manually enter the floor information to continue navigation, or switching to the floor navigation method based on the map.

[0151] In the embodiment of the present invention, measure the barometric pressure on different floors inside the building and record the barometric pressure value corresponding to each floor. Organize the measured barometric pressure values and the corresponding floor information to establish a mapping relationship table between barometric pressure and floors, that is, a barometric pressure - floor mapping table. Obtain the barometric pressure data at the current location through a barometric pressure measurement device, match the obtained barometric pressure data with the barometric pressure - floor mapping table, find the floor corresponding to the closest barometric pressure value, and determine it as the initial floor. Receive and parse the destination floor information input by the user. According to the initial floor and the target floor, combined with the floor layout and passage information inside the building, plan a navigation path from the current floor to the target floor. During the navigation process, continuously monitor the barometric pressure data at the current location through the barometric pressure measurement device. Compare the monitored barometric pressure data with the barometric pressure - floor mapping table. If it is found that the current floor information is different from the previous matching result, update the floor information. Re-plan the navigation path according to the updated floor information to ensure the accuracy and real-time nature of the navigation. Design an intuitive and easy-to-use floor navigation interface to display information such as the current floor, target floor, and path planning. Provide real-time floor navigation guidance to the user through visual (such as floor name display, distance information prompt, etc.) and voice (such as turning prompt, floor switching prompt, etc.) means. When the user reaches the target floor, issue an arrival prompt through visual and voice means and end the navigation process. During the navigation process, if abnormal situations such as barometric pressure measurement device failure and floor information mismatch are encountered, handle them according to the preset abnormal handling mechanism, such as prompting the user to manually enter the floor information to continue navigation, or switching to the floor navigation method based on the map.

[0152] Suppose an intelligent navigation device is used for floor navigation in an office building.

[0153] The device measures the air pressure on different floors inside the building, records the air pressure values corresponding to each floor, and establishes an air pressure - floor mapping table. The user activates the navigation device and inputs the destination floor as the 5th floor. The device obtains the air pressure data at the current location and matches it with the air pressure - floor mapping table to determine that the initial floor where the user is located is the 2nd floor. Based on the initial floor (2nd floor) and the target floor (5th floor), combined with the floor layout and passage information inside the building, the device plans a navigation route from the 2nd floor to the 5th floor. During the navigation process, the device continuously monitors the air pressure data at the current location. When the user takes the elevator from the 2nd floor to the 3rd floor, the device detects the change in air pressure and updates the floor information to the 3rd floor. Since the floor information has changed, the device re - plans the navigation route. The device provides real - time floor navigation guidance to the user through visual and voice means, including the current floor name (3rd floor), the distance information to the target floor (2 more floors to reach the 5th floor), and turning prompts (please go to the elevator entrance). When the user reaches the target floor (5th floor), the device gives an arrival prompt and ends the navigation process. If an abnormal situation occurs during the navigation process (such as a malfunction of the air pressure measuring device), the device prompts the user to manually input the floor information to continue the navigation according to the preset abnormal handling mechanism.

[0154] Through the mapping relationship between air pressure data and floor information, accurate positioning of floor navigation is achieved, improving the accuracy of navigation. Continuously monitoring the air pressure data during the navigation process and updating the floor information in real - time according to the change in air pressure ensure the real - time and accuracy of navigation. Providing real - time floor navigation guidance to the user through visual and voice means enhances the user experience and convenience. This floor navigation method does not depend on the specific layout or equipment inside the building, and only requires air pressure measurement and floor information recording, so it has strong adaptability and scalability. A preset abnormal handling mechanism can handle possible abnormal situations during the navigation process, ensuring the stability and reliability of navigation.

[0155] In a preferred embodiment of the present invention, the user interaction interface module 17, which is used to display three - dimensional navigation information, including horizontal position, height, floor information, route planning, and navigation instructions, may include:

[0156] In the embodiments of the present invention, the current location information (latitude and longitude, altitude) of the device is obtained through positioning systems such as GPS and Beidou, and the floor information is obtained by combining sensor data (such as barometers, accelerometers) and indoor positioning technologies (such as Wi-Fi positioning, Bluetooth beacons). The map service API is used for route planning to obtain the route information from the starting point to the ending point. 3D models of terrain, buildings, roads, etc. are created, or 3D data provided by the map service is used. The user's location marker and perspective are updated in real time according to the location data, and the floor information is displayed in the 3D model, which can be distinguished by different colors or labels. The planned route is drawn on the 3D map, and different colors or line styles are used to represent it. According to the current location and route information, real-time turning prompts (such as turn left at the next intersection) are provided, and text-to-speech synthesis technology is integrated to provide voice navigation instructions. Operations such as zooming, rotating, and panning of the map are implemented, multi-touch is supported, and easy-to-operate menus and buttons are designed for switching different functions (such as floor selection, route replanning), and detailed information such as the current location, destination, and estimated arrival time is displayed.

[0157] Embodiments of the present invention also provide a computing device, including: a processor, and a memory storing a computer program. When the computer program is run by the processor, it executes the system as described above. All implementation manners in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0158] Embodiments of the present invention also provide a computer-readable storage medium storing instructions. When the instructions are run on a computer, the computer is caused to execute the system as described above. All implementation manners in the above system embodiments are applicable to this embodiment and can achieve the same technical effects.

[0159] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A three-dimensional multi-source navigation information system based on Beidou, characterized in that, Including: A Beidou satellite navigation and positioning module, which is used to provide basic positioning information, including longitude, latitude, altitude and time data; A multi-sensor data acquisition module, including a barometer, an accelerometer and a thermometer, which is used to acquire environmental data, including barometric pressure value, acceleration change and temperature; A data fusion processing module, which is used to dynamically update and fuse the data of each sensor by using Bayesian maximum likelihood estimation and Kalman filtering, and smooth and optimize the positioning data through forward-backward combined filtering algorithm and smoothing algorithm to obtain a positioning result containing altitude and position information; A three-dimensional positioning module, which is used to realize centimeter-level altitude positioning by using barometric pressure change information according to the temperature and barometric pressure ranges of the longitude and latitude areas of the current position of the Beidou satellite navigation and positioning module; A three-dimensional navigation function module, which is used to realize horizontal navigation and vertical navigation in complex terrain, urban canyons and urban viaduct scenarios according to the positioning result of altitude and position information; An urban space stratification strategy module, which is used to stratify the urban space according to barometric pressure data and floor information, and match the barometric pressure data of the current position during navigation to determine the floor where it is located, so as to realize the floor navigation function; A user interaction interface module, which is used to display three-dimensional navigation information, including horizontal position, altitude, floor information, route planning and navigation instructions.

2. The Beidou-based three-dimensional multi-source navigation information system according to claim 1, characterized in that Using Bayesian maximum likelihood estimation and Kalman filtering, dynamically update and fuse the data of each sensor, and smooth and optimize the positioning data through forward-backward combined filtering algorithm and smoothing algorithm to obtain a positioning result containing altitude and position information, including: According to the prior information and the current observation data, use the Bayesian maximum likelihood estimation method to calculate the posterior probability distribution of each sensor data to obtain the final estimated value of each sensor data; According to the final estimated values of each sensor data, establish a forward Kalman filtering model, and through iterative calculation, obtain the forward filtering state vector and error covariance matrix; according to the forward filtering state vector, rewrite the backward filtering model to obtain the backward filtering state vector and error covariance matrix; At each moment k, use the forward filtering state vector and the backward filtering state vector to obtain the state vector and error covariance of the combined filtering through information fusion, and use the smoothing algorithm to perform backward smoothing on the combined filtering result to obtain the smoothed state vector and covariance matrix; the initial value of the backward smoothing is selected as the state vector and covariance matrix of the end epoch of filtering; Extract the position information, including longitude, latitude and altitude, from the smoothed state vector as the final positioning result.

3. The Beidou-based three-dimensional multi-source navigation information system according to claim 2, characterized in that, The calculation process of the posterior probability distribution of each sensor data includes: Based on Bayes' theorem, calculate the posterior probability distribution through the joint probability of the sensor observation data and the prior distribution information, where: For the observation data point of the i-th sensor, construct the numerator part of the exponential joint probability function as a weighted combination of the observation data deviation term and the parameter deviation term, where the observation data deviation term is the normalized squared difference of the variance between the current measurement value and the parameter-related expected mean, and the parameter deviation term is the normalized squared difference of the prior variance between the current parameter value and the prior mean; The denominator part of the posterior probability distribution is constructed as a product combination of a normalization integral factor, the observation variance, and the prior variance constant term, where the normalization integral factor realizes probability normalization through global integration over the parameter space; The variance of the observed data is calculated using the finite-sample unbiased estimation method. Specifically, the sample mean of the n measurements of the i-th sensor is calculated, and then the sum of the squared deviations of each measurement from the sample mean is divided by the degrees of freedom parameter. The prior mean and prior variance of the parameter are determined based on the historical calibration data of the sensor, where the prior mean represents the expected central value of the parameter, and the prior variance represents the range of uncertainty in the parameter estimation. The probability distribution calculation process finally outputs the posterior probability distribution of the parameter for each sensor under the given observed data, which is used for the iterative optimization of the Bayesian maximum likelihood estimation.

4. The Beidou-based three-dimensional multi-source navigation information system according to claim 3, characterized in that According to the temperature and pressure ranges in the latitude and longitude area of the current position of the Beidou satellite navigation and positioning module, centimeter-level altitude positioning is achieved using barometric pressure change information, including: Decoding the positioning signal of the Beidou satellite to obtain the latitude and longitude coordinates, speed, and time information of the current position, and obtaining the temperature and pressure ranges in the latitude and longitude area of the current position from the meteorological data source; Based on the relationship between barometric pressure and altitude, a standard atmosphere model is established, and the parameters of the standard atmosphere model are adjusted according to the changes in temperature and pressure; Calculating a preliminary altitude value based on the barometric pressure data of the current position; Using the known altitude point as the reference position, calculating the barometric pressure change rate between the current position and the reference position, and correcting the preliminary altitude value using the standard atmosphere model according to the barometric pressure change rate to obtain the corrected altitude value; Measuring the barometric pressure of the current position using a barometric pressure sensor and combining it with the corrected altitude value to achieve centimeter-level altitude positioning.

5. The Beidou-based three-dimensional multi-source navigation information system according to claim 4, wherein According to the positioning results of altitude and position information, horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios are achieved, including: Receiving the Beidou signal to obtain the altitude and position information of the current position, and using map data and real-time sensor information to identify the terrain type where the user is currently located. At the same time, combining the Beidou signal, positioning in three-dimensional space is achieved, including horizontal position and vertical altitude; According to the destination input by the user and the current position, combining map data and real-time traffic information to plan the final horizontal navigation path; During the navigation process, providing real-time navigation guidance to the user in visual and voice ways, including turning prompts, road names, and distance information, and dynamically adjusting the navigation path according to the real-time traffic conditions and road conditions; After navigation, according to the user's needs and the altitude information of the current position, planning the altitude change range of vertical navigation, and combining the Beidou signal and the data of the altitude sensor during the planning process to achieve horizontal and vertical navigation in complex terrain, urban canyons, and urban viaduct scenarios.

6. The Beidou-based three-dimensional multi-source navigation information system according to claim 5, characterized in that, According to the barometric pressure data and floor information, the urban space is stratified, and the barometric pressure data of the current position is matched during the navigation process to determine the floor where the user is located, realizing the floor navigation function, including: According to the air pressure data inside the building, by measuring the air pressure on different floors and recording the corresponding floor information, a mapping relationship between air pressure and floors is established to obtain an air pressure - floor mapping table; When the navigation starts, obtain the air pressure data of the current location, and match the current air pressure data with the air pressure - floor mapping table to determine the initial floor where the user is located; According to the destination floor input by the user and the currently matched initial floor, plan the navigation path from the current floor to the target floor; During the execution of the navigation, continuously monitor the air pressure data of the current location, and update the floor information in real time according to the air pressure change. If the current floor information is different from the matched result, re - plan the navigation path according to the new floor information; Provide real - time floor navigation guidance to the user in visual and voice ways, including the current floor name, the distance information from the target floor, and turning prompts. The user moves between floors according to the floor navigation guidance; When the user arrives at the target floor, issue an arrival prompt and end the navigation process. If an abnormal situation occurs during the navigation, handle it according to the preset abnormal handling mechanism, including prompting the user to manually input the floor information to continue the navigation, or switching to the floor navigation method based on the map.

Images