Mining navigation resolving method based on reverse estimation and forward compensation

Abstract

The invention provides a mining navigation resolving method based on backstepping and forward compensation, and relates to the technical field of mine intellectualization, the method comprises the following steps: using external equipment, including a total station, a laser radar and a target, to obtain displacement point data of an end point of a current path section; converting the displacement point data of the terminal point of the current path section into position information including longitude, latitude and height; according to the position information subjected to latitude and longitude conversion, the attitude matrix and the speedometer incremental data, the current path is reversely calculated. According to the method, the displacement data is obtained through the external equipment, the position information is converted, the sampling point position is calculated in combination with the multi-source data, and compensation is performed after comparison with the original data and error evaluation, so that continuous high-precision correction of the navigation data is realized, the problem of error accumulation is effectively solved, and the accuracy and stability of the mining navigation system are improved.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent mines, and in particular to a mine navigation solution method based on reverse thrust and forward compensation. Background Art

[0002] Inverse calculation is performed for each segment, and system errors need to be estimated and compensated, which increases the complexity of the calculation. In a mining environment with high real-time requirements, this high computational complexity may cause delays in navigation information and even affect the overall performance of the navigation system. The effectiveness of reverse calculation and forward compensation depends largely on the accuracy and reliability of external sensors. However, in a special environment such as a mine, external sensors may be subject to various interferences, such as electromagnetic interference, occlusion, dust, etc., which may cause their performance to degrade or fail, thereby affecting navigation accuracy.

[0003] Although reverse and forward compensation can reduce the error within each segment, the error may still be transmitted and accumulated between segments. Especially when there are a large number of segments or the navigation time is long, this error transmission and accumulation may affect the final navigation results. The mine environment is complex and changeable, and there are various uncertainties, such as mine car bumps, tunnel deformation, electromagnetic interference, etc. These factors may have an adverse effect on the performance of the navigation system, and the traditional reverse and forward compensation methods may lack sufficient adaptability in the face of these uncertainties, resulting in reduced navigation accuracy or system failure.

[0004] In order to implement reverse thrust and forward compensation, high-precision sensors and complex processing algorithms are required, which increases the cost of the system. At the same time, due to the harsh environment of the mine, these sensors and equipment may require frequent maintenance and calibration to ensure their accuracy and reliability, which also increases maintenance costs. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a mining navigation solution method based on reverse calculation and forward compensation, which can effectively reduce the error of the whole process solution.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] In the first aspect, a mining navigation solution method based on reverse calculation and forward compensation comprises:

[0008] Using external equipment, including total station, laser radar and target, to obtain the displacement point data of the end point of the current path segment;

[0009] Convert the displacement point data of the end point of the current path segment into position information including longitude, latitude and altitude to obtain the position information after longitude and latitude conversion;

[0010] Reversely calculate the current path based on the position information, attitude matrix and odometer increment data after longitude and latitude conversion;

[0011] The position information of each reverse sampling point is compared with the original data of the corresponding strapdown inertial navigation odometer integrated navigation, and the error of the system is evaluated and estimated through filtering to obtain the error estimation result;

[0012] According to the result of error estimation, error compensation is performed on the navigation data of the next section, and the process of reverse deduction, estimation and compensation is repeated to achieve continuous correction and estimation of the navigation data of the mining navigation system.

[0013] Furthermore, the location information after longitude and latitude conversion is p n =[L n λ n h n ] T ; Among them, p n Indicates the location information of the end point of the nth path; L n Indicates the latitude of the end point of the nth path; λ n Indicates the longitude of the end point of the nth path; h n represents the height of the end point of the nth path; T represents the transpose of the vector.

[0014] Furthermore, the current path is calculated inversely based on the position information, attitude matrix and odometer increment data after latitude and longitude conversion, including:

[0015] Determine the end point position p of the nth path n , including latitude L n , longitude λ n and height h n , and obtain the attitude matrix, which contains the attitude information of the navigation system in three axes, namely heading, pitch and roll;

[0016] Obtain odometer incremental data, including the distance and direction changes of the vehicle in each sampling time interval, and set the starting point of the reverse calculation as the end point position p n , which is the end point p of the nth path nm ;

[0017] Initialize an empty list to store the position information of each sampling point calculated during the inverse process;

[0018] Starting from the end point, traverse each sampling time point t in the nth segment of the path in reverse i , where i traverses from m to 1, and m represents the total number of sampling points in the nth segment;

[0019] For each sampling time point t i, using the attitude matrix and odometer increment data to calculate from the current sampling point p nm To the next sampling point, t i -1 displacement vector Δp i ;

[0020] For each calculated displacement vector Δp i , applied to the current position, to obtain the position coordinates p of the next sampling point i ′;

[0021] When i=m-1, Among them, P' n(m-1) Represents the position coordinates of the m-1th sampling point on the nth path obtained by reverse calculation; represents the attitude matrix; D represents the distance part of the odometer increment data; R nm Represents the rotation matrix associated with the time interval from the mth sampling point to the m-1th sampling point;

[0022] For each i, the position coordinates are calculated and updated using the attitude matrix, odometer increment data, and rotation matrix, and each calculated position coordinate p i 'Add to storage list;

[0023] After the reverse calculation is completed, the storage list contains the nth segment path from the starting position to the end position p nm The location information of each sampling point.

[0024] Furthermore, the posture matrix for:

[0025]

[0026] Among them, cosψ and sinψ represent the cosine and sine values ​​of the yaw angle ψ; cosθ and sinθ represent the cosine and sine values ​​of the pitch angle θ; cosφ and sinφ represent the cosine and sine values ​​of the roll angle φ.

[0027] Furthermore, the position information of each reverse sampling point is compared with the original data of the corresponding strapdown inertial navigation odometer integrated navigation, and the system error is evaluated and estimated through filtering processing to obtain the error estimation results, including:

[0028] Obtain the original data of the nth strapdown inertial navigation odometer integrated navigation, including the position information of each sampling time point, and obtain the position information of each sampling point;

[0029] For each sampling point, the position difference between the two sets of data is calculated by Euclidean distance, and the position difference of each sampling point is formed into a position difference data set;

[0030] Performing filtering processing on the position difference data set to obtain filtered position difference data;

[0031] According to the position difference data after filtering, the error of the system is evaluated, including the statistical indicators of the mean, standard deviation, maximum and minimum values ​​of the error are calculated to obtain the result of the error evaluation;

[0032] The error evaluation results are analyzed through the error model to obtain the error estimation results.

[0033] Furthermore, the error evaluation results are analyzed through the error model to obtain the error estimation results, including:

[0034] Construct an error model and set the initial parameters of the error model;

[0035] The error evaluation results, including the error mean and standard deviation of each sampling point, are input into the error model for training to obtain a trained error model;

[0036] The error model after training is used to analyze the error, including the source, type, and distribution characteristics of the error, and the error estimation result is obtained, including the size, direction, and distribution information of the error.

[0037] Furthermore, the calculation formula of Euclidean distance is:

[0038]

[0039] Where, ΔE i represents the weighted position difference of the i-th sampling point; w x 、w y 、w z represents the weight coefficient; Represents the coordinate values ​​of the theoretical position of the i-th measurement point in the x, y, and z directions; At time τ i When , the actual measurement position of the i-th measurement point; b x 、b y 、b z Indicates the offset in the x, y, and z directions; Represents the noise variance of the ith measurement point in the x, y, and z directions.

[0040] Furthermore, according to the result of the error estimation, the error compensation is performed on the navigation data of the next segment, and the continuous correction and estimation of the navigation data of the mining navigation system is realized by continuously repeating the process of reverse deduction, estimation and compensation, including:

[0041] According to the error estimation result, the navigation data of the n+1th segment is compensated for the error, including adjusting the position information, correcting the attitude matrix, and calibrating the odometer to obtain the compensated navigation data;

[0042] The compensated navigation data is used as input data for the next iteration, and the compensated navigation data is continuously used as a new starting point to continuously correct and estimate the navigation data.

[0043] Secondly, the mining navigation solution system based on reverse calculation and forward compensation includes:

[0044] Data acquisition module, used to collect navigation data in the mine in real time, including position information, speed information, direction information and environmental parameters;

[0045] The reverse calculation module is used to reversely calculate the movement trajectory of the mine car based on the navigation data of the data acquisition module to obtain the previous position of the mine car;

[0046] The forward compensation module is used to perform forward compensation on the reverse calculation results according to the geographical information of the mine, so as to correct the error caused by the mine environment and obtain the current position of the mine car;

[0047] A map matching module is used to match the compensated mine car position with the mine map to obtain a map matching result;

[0048] The navigation solution module is used to perform navigation solution based on the results of inverse calculation, forward compensation and map matching.

[0049] According to a third aspect, a computing device includes:

[0050] one or more processors;

[0051] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0052] The above solution of the present invention includes at least the following beneficial effects:

[0053] The displacement point data of the end point of the path is accurately obtained through external equipment (total station, lidar and target) and converted into accurate position information. Combining these position information, attitude matrix and odometer increment data, the position information of each sampling point on each path segment can be calculated more accurately. Subsequently, by inverse calculation and comparison of the original data, the system error is accurately evaluated and estimated, and then the error compensation of subsequent navigation data is performed, thereby improving the accuracy of the entire navigation process. The accumulation of errors is a serious problem during long-term navigation. By regularly using external equipment for position correction, combined with inverse calculation and forward compensation technology, the accumulated errors can be corrected in time at the end of each segment to ensure that the starting data of the next navigation segment is accurate and reliable. This mechanism effectively suppresses the accumulation of errors and improves the long-term stability of navigation data.

[0054] Traditional multi-source information fusion methods rely on real-time data from all sensors, but external devices may be restricted or interfered by the environment. This method allows data fusion and correction under conditions where the external device is in a good environment, reducing dependence on external devices and improving system reliability. Even in scenarios where some external devices cannot be used, the system can still maintain a certain navigation capability.

[0055] By repeatedly repeating the process of reverse deduction, estimation and compensation, this method can realize the continuous correction and estimation of the navigation data of the mining navigation system. This ensures the continuity and real-time nature of the navigation data, and provides timely and accurate navigation information for the automation and intelligent operation of the mine. Instead of relying entirely on external equipment for real-time navigation, external equipment is used for correction and estimation when necessary. This flexible use method optimizes the utilization of resources, reduces the dependence on high-precision external equipment, and thus reduces the overall cost of the system. Accurate navigation information is the key to the efficiency and safety of mine operations. By improving navigation accuracy and stability, it can help mines achieve more accurate mining, transportation and monitoring, thereby improving operating efficiency, reducing safety risks, and providing a strong guarantee for the intelligent development of mines. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flow chart of a mining navigation solution method based on inverse calculation and forward compensation provided by an embodiment of the present invention.

[0057] Figure 2 It is a schematic diagram of a mining navigation solution system based on inverse calculation and forward compensation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying 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 to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0059] like Figure 1 As shown, an embodiment of the present invention proposes a mining navigation solution method based on reverse calculation and forward compensation, and the method includes the following steps:

[0060] Step 11, using external equipment, including a total station, a laser radar, and a target, to obtain the displacement point data of the end point of the current path segment;

[0061] Step 12, converting the displacement point data of the end point of the current path segment into position information including longitude, latitude and altitude, and obtaining position information after longitude and latitude conversion;

[0062] Step 13, reversely calculate the current path based on the position information after longitude and latitude conversion, the attitude matrix and the odometer increment data;

[0063] Step 14, comparing the position information of each reverse sampling point with the original data of the corresponding strapdown inertial navigation odometer integrated navigation, evaluating and estimating the system error through filtering processing, and obtaining the error estimation result;

[0064] Step 15, based on the result of error estimation, error compensation is performed on the navigation data of the next segment, and the process of reverse deduction, estimation and compensation is repeated continuously to achieve continuous correction and estimation of the navigation data of the mining navigation system.

[0065] In an embodiment of the present invention, the precise displacement point data of the end point of the path is obtained by external high-precision equipment (such as a total station, a laser radar, etc.), thereby ensuring the accuracy of the navigation starting point. The displacement point data is converted into a standardized position information format (longitude, latitude, altitude), thereby improving the versatility and readability of the data. Combining the position information, the attitude matrix and the odometer increment data, the position information of each sampling point on the path can be accurately calculated, thereby improving the accuracy and delicacy of navigation. By comparing the calculated sampling point position information with the original data of the strapdown inertial navigation odometer combined navigation, and by filtering, the error of the system can be accurately evaluated and estimated, thereby effectively improving the accuracy and stability of the navigation system. According to the result of the error estimation, the error compensation is performed on the next section of the navigation data, thereby eliminating or reducing the error generated in the previous section of the navigation process, thereby ensuring the continuity and accuracy of the navigation data. By continuously repeating the process of inverse deduction, estimation and compensation, the continuous correction and estimation of the navigation data of the mining navigation system can be achieved, thereby improving the long-term stability and reliability of the navigation system.

[0066] In a preferred embodiment of the present invention, in the above step 11, external equipment including a total station, a laser radar and a target are used to obtain the displacement point data of the end point of the current path segment; in the above step 12, the displacement point data of the end point of the current path segment is converted into position information including longitude, latitude and altitude to obtain the position information after longitude and latitude conversion; the position information after longitude and latitude conversion is p n =[L n λ n h n ] T ; Among them, p n Indicates the location information of the end point of the nth path; L n Indicates the latitude of the end point of the nth path; λ n Indicates the longitude of the end point of the nth path; h n Indicates the height of the end point of the nth path; T represents the transpose of the vector, which can include:

[0067] Ensure that the total station is properly installed and calibrated, and can stably measure the three-dimensional coordinates of the target point. Select a lidar with high-precision distance and angle measurement capabilities, and ensure that it can accurately scan and identify the target. Prepare a sufficient number of targets and ensure that they have clear markings and stable placement so that the total station and lidar can accurately identify and measure. Place the target at the end of the nth path, and ensure that the target is in a stable position and is not blocked.

[0068] Start the total station and set its measurement mode and parameters. Aim the target at the end of the nth path through the telescope of the total station and make precise measurements. The total station will record the three-dimensional coordinates of the target (including longitude, latitude and altitude), which are the required displacement point data.

[0069] Start the LiDAR and set its scanning range and resolution. The LiDAR starts scanning and captures 3D point cloud data of the surrounding environment. Identify and extract the point cloud data of the target at the end of the nth path in the scanned data. By processing the point cloud data, the displacement point data measured by the total station can be further verified and corrected. Summarize and process the data obtained by the total station and LiDAR measurements. If measurements are taken using multiple devices, data fusion and verification are required to ensure data accuracy and consistency. Convert the processed data into a standardized location information format, such as longitude, latitude, and altitude [L n λ n h n ].

[0070] Represent these data in vector form, namely p n =[L n λ n h n ] T , where T represents the transpose of the vector. The acquired displacement point data is stored in the database of the navigation system.

[0071] In a preferred embodiment of the present invention, the above step 13, based on the position information converted from longitude and latitude, the attitude matrix and the odometer increment data, reversely calculates the current path, which may include:

[0072] Step 121, determine the end point position p of the nth path n , including latitude L n , longitude λ n and height h n , and obtain the attitude matrix, which contains the attitude information of the navigation system in three axes, namely heading, pitch and roll;

[0073] Step 122, obtain the odometer increment data, including the distance and direction change of the vehicle in each sampling time interval, and set the starting point of the reverse calculation as the end point position p n , which is the end point p of the nth path nm ;

[0074] Step 123, initializing an empty list for storing the position information of each sampling point calculated in the reverse process;

[0075] Step 124, starting from the end point, reversely traverse each sampling time point t in the nth segment path i , where i traverses from m to 1, and m represents the total number of sampling points in the nth segment;

[0076] Step 125, for each sampling time point t i, using the attitude matrix and odometer increment data to calculate from the current sampling point p nm To the next sampling point, t i -1 displacement vector Δp i ;

[0077] Step 126, for each calculated displacement vector Δp i , applied to the current position, to obtain the position coordinates p of the next sampling point i ′;

[0078] Step 127, when i=m-1, Among them, P' n(m-1) Represents the position coordinates of the m-1th sampling point on the nth path obtained by reverse calculation; represents the attitude matrix; D represents the distance part of the odometer increment data; R nm Represents the rotation matrix associated with the time interval from the mth sampling point to the m-1th sampling point;

[0079] Step 128, for each i, calculate and update the position coordinates using the attitude matrix, odometer increment data and rotation matrix, and update each calculated position coordinate p i 'Add to the storage list;

[0080] Step 129: After the reverse calculation is completed, the storage list contains the path from the starting position to the end position p on the nth segment. nm The location information of each sampling point.

[0081] In the embodiment of the present invention, the location data of the end point of the nth path obtained from an external device (such as a total station, a laser radar, etc.) includes the latitude L n , longitude λ n and height h n , and represented as vector p n =[L n λ n h n ] T At the same time, the attitude matrix of the navigation system is obtained, which contains the attitude information of the navigation system in three axes, namely heading, pitch and roll.

[0082] Step 122, obtain the distance and direction change of the vehicle in each sampling time interval from the vehicle's odometer sensor. Set the starting point of the reverse calculation to the end point position p n , which is the end point P of the nth path nm .

[0083] Step 123: Create an empty list to store the position information of each sampling point calculated during the inverse process.

[0084] Step 124, starting from the end point, reversely traverse each sampling time point t in the nth segment path i , where i traverses from m to 1, and m represents the total number of sampling points in the nth segment.

[0085] Step 125, for each sampling time point t i , using the attitude matrix and odometer increment data to calculate from the current sampling point P nm To the next sampling point (ie t i -1) displacement vector Δp i The calculation of the displacement vector needs to take into account the distance traveled by the vehicle, the change in direction, and the heading, pitch, and roll information in the attitude matrix.

[0086] Step 126, for each calculated displacement vector Δp i , apply it to the current position P nm , get the position coordinates p of the next sampling point i ′. This step is achieved through vector addition, that is, p i ′=P nm -Δp i (Note that the direction is reverse, so use subtraction).

[0087] Step 127, when i=m-1, that is, when processing the first inverse point, the rotation matrix R needs to be specially considered nm Calculate Among them, P' n(m-1) Represents the position coordinates of the m-1th sampling point on the nth path obtained by reverse calculation; represents the attitude matrix; D represents the distance part of the odometer increment data; R nm Represents the rotation matrix associated with the time interval from the mth sampling point to the m-1th sampling point. This step is to more accurately consider the direction change of the vehicle during movement.

[0088] Step 128, for each i, calculate and update the position coordinate p using the attitude matrix, odometer increment data and rotation matrix i ′. Each calculated position coordinate p i ’ is added to the previously initialized storage list.

[0089] Step 129: After the reverse calculation is completed, the storage list contains the path from the starting position to the end position P on the nth segment. nm The position information of each sampling point can be used for navigation solution, error evaluation and compensation, etc.

[0090] The acquisition of the attitude matrix makes it possible to consider the change in vehicle direction and improves the accuracy of the reverse calculation. The acquisition of odometer incremental data provides the necessary basis for calculating the position of each sampling point. The design of reverse traversal makes the reverse process more consistent with the actual way of obtaining navigation data. Through reverse traversal and calculation of displacement vectors, the position information of each sampling point on the nth path can be accurately obtained. Special treatment of the first reverse point takes into account the influence of the rotation matrix, which further improves the accuracy of the reverse calculation. The use of storage lists makes the reverse results more orderly and easy to handle. After the reverse calculation is completed, the obtained position information can be used for subsequent navigation solution and error evaluation, which makes it possible for continuous correction and estimation of the mining navigation system. The entire reverse process takes into account the vehicle's travel distance, direction change and attitude information, making the reverse result closer to the actual situation and improving the accuracy and reliability of the navigation system.

[0091] In another preferred embodiment of the present invention, the posture matrix for:

[0092]

[0093] Among them, cosψ and sinψ represent the cosine and sine values ​​of the yaw angle ψ; cosθ and sinθ represent the cosine and sine values ​​of the pitch angle θ; cosφ and sinφ represent the cosine and sine values ​​of the roll angle φ.

[0094] In the embodiment of the present invention, the posture matrix It is obtained by multiplying three rotation matrices, corresponding to the yaw angle (heading angle) ψ, pitch angle θ and roll angle φ. Each rotation matrix is ​​a 3x3 orthogonal matrix, representing the rotation around an axis (X, Y or Z axis). The yaw angle ψ represents the rotation of the navigation system around the vertical axis (Z axis).

[0095] The corresponding rotation matrix is This matrix rotates the coordinate system by angle ψ around the Z axis. The pitch angle θ represents the rotation of the navigation system around the Y axis. The corresponding rotation matrix is

[0096]

[0097] This matrix rotates the coordinate system by an angle θ around the Y axis. The roll angle φ represents the rotation of the navigation system around the X axis. The corresponding rotation matrix is This matrix rotates the coordinate system by angle φ around the X axis.

[0098] Multiply the above three rotation matrices in the order of yaw angle, pitch angle, and roll angle to obtain the attitude matrix During the inverse process, the pose matrix is ​​used to calculate the displacement vector for each sample point, taking into account the change in orientation of the vehicle.

[0099] The attitude matrix can accurately convert the position and orientation of an object in the local coordinate system (the coordinate system associated with the object itself) into the global or navigation coordinate system. In areas where precise attitude control is required, the attitude matrix provides an accurate description of the current attitude of the object, allowing the control system to adjust the attitude of the object as needed to keep it stable or perform specific actions. The use of the attitude matrix simplifies complex three-dimensional space calculations. Through matrix operations, geometric transformations such as rotation and translation can be easily processed without tedious trigonometric calculations or manual adjustment of coordinate systems.

[0100] In a preferred embodiment of the present invention, the above step 14, comparing the position information of each reverse sampling point with the corresponding original data of the strapdown inertial navigation odometer integrated navigation, evaluating and estimating the error of the system through filtering processing, and obtaining the result of error estimation, may include:

[0101] Step 141, obtaining the original data of the nth strapdown inertial navigation odometer integrated navigation, including the position information of each sampling time point, and obtaining the position information of each sampling point;

[0102] Step 142, for each sampling point, calculate the position difference between the two sets of data by using the Euclidean distance, and form a position difference data set with the position difference of each sampling point;

[0103] Step 143, filtering the position difference data set to obtain filtered position difference data;

[0104] Step 144, evaluating the error of the system according to the position difference data after filtering, including calculating the mean, standard deviation, maximum and minimum statistical indicators of the error, and obtaining the result of error evaluation;

[0105] Step 145, analyzing the error evaluation result through the error model to obtain the error estimation result.

[0106] In an embodiment of the present invention, raw data is read from a strapdown inertial navigation odometer integrated navigation system, and these data include longitude, latitude and altitude information at each sampling time point. At the same time, the location information of each sampling point obtained by other methods (such as GPS, map matching, etc.) is obtained as reference data. Ensure that the timestamps of the two sets of data match. Use an appropriate data interface to extract data. Preprocess the data, such as removing outliers, filling missing values, etc. Arrange the data into a format that is easy to handle, such as an array or a data frame.

[0107] Step 142, for two given sets of data (each set of data contains the longitude, latitude and altitude of the sampling point), calculate the position difference of each sampling point in the two sets of data, and record these differences to form a position difference data set. The position difference is measured by Euclidean distance, and the longitude and latitude need to be converted into x, y, z coordinates in the Cartesian coordinate system before calculation.

[0108] Specifically, ensure that the timestamps or indices of the two sets of data match exactly so that each sampling point can be accurately mapped. Extract the longitude, latitude, and altitude information of each sampling point in each set of data. For each sampling point, use the following formula to convert the longitude and latitude to the x, y, and z coordinates in the Cartesian coordinate system:

[0109] x=R·cos(L n )·cos(λ n )

[0110] y=R·cos(L n )·sin(λ n )

[0111] z=R·sin(L n )

[0112] Where R is the radius of the earth, which is 6371 kilometers. For each sampling point, the Euclidean distance formula is used to calculate the position difference between the two sets of data. The position difference value of each sampling point is recorded to form a position difference data set. This data set can be a simple list or table, in which each element or row contains the position difference value of the corresponding sampling point. Euclidean distance is a measure of the straight-line distance between two points. In three-dimensional space, it considers the differences in the three directions of x, y, and z. By calculating the Euclidean distance of each sampling point in the two sets of data, the position difference between the two sets of data at the sampling point can be obtained. The position difference data set is a list in which each element is the position difference value of the corresponding sampling point. The position difference data set is filtered to remove noise and outliers. Select appropriate filtering parameters, such as window size, filter coefficient, etc., to achieve the best filtering effect. Obtain the position difference data after filtering. Select the appropriate filtering method according to the data characteristics and application requirements.

[0113] Step 144, based on the position difference data after filtering, calculate statistical indicators such as the mean, standard deviation, maximum value and minimum value of the error. These statistical indicators can reflect the overall situation and fluctuation range of the system error, and obtain the result of error evaluation.

[0114] Step 145, based on the error evaluation results, establish an error model. The error model is linear. Use the error model to estimate and predict the error. Analyze the error source and influencing factors, and propose improvement suggestions or measures. Obtain the error estimation results. Select a suitable error model for estimation. Verify and calibrate the error model to ensure its accuracy and reliability.

[0115] By comparing the position information from different sources, we can get a preliminary understanding of the performance of the system. The position difference between the two sets of data is quantified, which provides a direct basis for evaluating the system error. Noise and outliers are effectively removed, which improves the accuracy and reliability of the position difference data. The filtered data is smoother and more continuous, which helps to evaluate the system error more accurately. By calculating statistical indicators, the overall situation and fluctuation range of the system error are comprehensively evaluated. An error model is established, which can estimate and predict the error, improving the reliability and accuracy of the system. By analyzing the source of the error and the influencing factors, targeted suggestions or measures are provided for the optimization and improvement of the system. The results of error estimation can provide important references for system performance evaluation, fault diagnosis and maintenance decisions.

[0116] In another preferred embodiment of the present invention, the above step 145, analyzing the error evaluation result through the error model to obtain the error estimation result, may include:

[0117] Step 1451, construct an error model and set initial parameters of the error model;

[0118] Step 1452, input the error evaluation results, including the error mean and standard deviation of each sampling point, into the error model for training to obtain a trained error model;

[0119] Step 1453, using the trained error model to analyze the error, including analyzing the source, type, and distribution characteristics of the error, and obtaining an error estimation result, including the size, direction, and distribution information of the error.

[0120] In an embodiment of the present invention, a suitable error model, such as a linear model, is selected according to the characteristics of the error and the application scenario. According to the type of the error model, the initial parameters that need to be set are determined. These parameters include coefficients, bias terms, standard deviations, mean values, etc. of the model. The setting of the initial parameters can be determined based on historical data. A programming language is used to construct a framework for the error model and set the initial parameters. Ensure that the model framework can receive input data (such as the results of error evaluation) and output the trained model.

[0121] Step 1452, organize the results of the error evaluation, including the error mean and standard deviation of each sampling point. Ensure that the format and structure of the data match the requirements of the error model. Perform necessary preprocessing on the input data, such as removing outliers, normalization, standardization, etc., to improve the training effect and accuracy of the model. Input the preprocessed input data into the error model for training. During the training process, the model will adjust its parameters according to the input data to minimize the error or achieve other optimization goals. Monitor the training process to ensure that the model converges and achieves the expected performance indicators. After the training is completed, save the trained error model.

[0122] Step 1453, load the previously saved trained error model. Use the trained error model to perform error analysis on new error data or original data. This can include calculating the predicted value of the error, analyzing the source, type and distribution characteristics of the error, etc. Based on the output of the model, extract the size, direction and distribution information of the error. Present the results of the error analysis in a chart, report or other form for easy understanding and analysis. The results should include key information such as the estimated value of the error, the distribution diagram of the error, the proportion of the error source, etc.

[0123] By building and training the error model, the size, direction and distribution of the error can be estimated more accurately, thereby improving the accuracy of positioning, measurement or prediction. The error model can help identify the main sources and types of errors, such as systematic errors, random errors, measurement errors, etc., and provide strong support for subsequent error correction and optimization. By analyzing the distribution characteristics and sources of errors, the system can be optimized and improved in a targeted manner, such as adjusting sensor parameters, improving algorithms, etc., to improve the overall performance and stability of the system. Accurate error estimation results can provide an important basis for decision making. For example, in a navigation system, error estimation results can help determine the final path, evaluate positioning accuracy, etc.

[0124] In a preferred embodiment of the present invention, the calculation formula of the Euclidean distance is:

[0125]

[0126] Where, ΔE i represents the weighted position difference of the i-th sampling point; w x 、w y 、w z represents the weight coefficient; Represents the coordinate values ​​of the theoretical position of the i-th measurement point in the x, y, and z directions; At time τ i When , the actual measurement position of the i-th measurement point; b x 、b y 、b zIndicates the offset in the x, y, and z directions; Represents the noise variance of the ith measurement point in the x, y, and z directions.

[0127] In the embodiment of the present invention, the input data is determined, including the theoretical position coordinates of each sampling point Each sampling point is measured at the actual time τ i The actual measured position coordinates at The offset in the x, y, and z directions (b x , b y , b z ). The noise variance of each sampling point in the x, y, and z directions Weight coefficient (w x , w y , w z ). For each sampling point, initialize a variable to store the weighted position difference (ΔE i ).

[0128] For each sampling point i, the following steps are performed:

[0129] Calculate the difference in the x direction Calculate the difference in the y direction Calculate the difference in the z direction Calculate the weighted position difference ΔE based on the weight coefficient and the difference i .

[0130] The formula takes into account the difference between the theoretical position and the actual measured position, and adjusts the importance of the difference in different directions (x, y, z) through weight coefficients. This provides a more precise and comprehensive quantification of position differences, which helps to more accurately evaluate the performance of the measurement system. Noise Variance The term allows the calculation to take into account the noise effect during the measurement process. x , w y , w z ) allows the weights of differences in different directions to be flexibly adjusted according to specific application scenarios. For example, in some applications, the position difference in the x direction may be more important than the difference in the y or z direction, so the weights of differences in different directions can be adjusted by increasing w x The value of t is used to reflect this. x , b y , b z ) allows the measurement results to be corrected to remove systematic offset errors.

[0131] In a preferred embodiment of the present invention, the above step 15, based on the result of the error estimation, performs error compensation on the navigation data of the next segment, and continuously repeats the process of inverse deduction, estimation and compensation to achieve continuous correction and estimation of the navigation data of the mining navigation system, which may include:

[0132] Step 151, according to the result of error estimation, error compensation is performed on the navigation data of the n+1th segment, including adjusting the position information, correcting the attitude matrix, and calibrating the odometer to obtain compensated navigation data;

[0133] Step 152, using the compensated navigation data as input data for the next iteration, and continuously using the compensated navigation data as a new starting point to continuously correct and estimate the navigation data.

[0134] In an embodiment of the present invention, the error estimation result of the n+1th segment of navigation data is obtained from the processing of the navigation data of the previous segment (nth segment). This includes position error, attitude error, and possible odometer error, etc. According to the position error in the error estimation, the position information in the n+1th segment of navigation data is adjusted. This involves fine-tuning the longitude, latitude, altitude or three-dimensional coordinates (x, y, z) to ensure that they more accurately reflect the actual position. Using the attitude error estimation, the matrix representing the attitude of the vehicle or device in the navigation data is corrected. The attitude matrix is ​​used to describe the direction and rotation state of an object in three-dimensional space. If the error estimate contains an odometer error, the odometer reading needs to be calibrated. This involves adjusting the offset of the odometer to ensure that it more accurately reflects the actual moving distance of the vehicle or device. The adjusted position information, the corrected attitude matrix and the calibrated odometer data are integrated to form compensated navigation data. These data will be used as input for the next iteration.

[0135] Step 152, sets the compensated navigation data as the starting point of the next iteration. This means that in subsequent iterations, these corrected data will be used as the new benchmark. Repeat the inverse, error estimation and compensation process in step 151 for the new navigation data segment (i.e., segment n+2 and subsequent segments). This includes obtaining the error estimation results again, adjusting the position information, correcting the attitude matrix, calibrating the odometer, and generating new compensated navigation data. Repeat the above process until the required navigation accuracy or the upper limit of the number of iterations is achieved. Through continuous correction and estimation, the errors in the navigation data can be gradually reduced and the accuracy of the navigation system can be improved.

[0136] By continuously correcting and estimating the navigation data, the position error and attitude error can be reduced, thereby improving the overall accuracy of the navigation system. This is particularly important for mining navigation systems, as accurate positioning and navigation are key to ensuring safe and efficient mining operations. The error compensation process helps to eliminate the accumulated errors in the navigation system, thereby enhancing the stability of the system. As the number of iterations increases, the navigation data will gradually approach the true value, reducing the system instability caused by error accumulation. Accurate navigation data can provide a better user experience. In mining navigation systems, this means that operators can more easily locate themselves, plan paths, and avoid obstacles, thereby improving work efficiency and safety. Continuously corrected and estimated navigation data provides the basis for the implementation of advanced navigation functions such as automatic path planning, obstacle avoidance navigation, etc. These functions rely on accurate navigation data to make decisions and perform tasks. By improving navigation accuracy and stability, equipment damage, accidents or downtime caused by navigation errors can be reduced. This will help reduce the maintenance cost of mining navigation systems and improve overall economic benefits.

[0137] like Figure 2 As shown, the embodiment of the present invention also provides a mining navigation solution system based on reverse calculation and forward compensation, including:

[0138] Data acquisition module, used to collect navigation data in the mine in real time, including position information, speed information, direction information and environmental parameters;

[0139] The reverse calculation module is used to reversely calculate the movement trajectory of the mine car based on the navigation data of the data acquisition module to obtain the previous position of the mine car;

[0140] The forward compensation module is used to perform forward compensation on the reverse calculation results according to the geographical information of the mine, so as to correct the error caused by the mine environment and obtain the current position of the mine car;

[0141] A map matching module is used to match the compensated mine car position with the mine map to obtain a map matching result;

[0142] The navigation solution module is used to perform navigation solution based on the results of inverse calculation, forward compensation and map matching.

[0143] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0144] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0145] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A mining navigation solution method based on reverse calculation and forward compensation is characterized in that: The method comprises: Using external equipment, including total station, laser radar and target, to obtain the displacement point data of the end point of the current path segment; Convert the displacement point data of the end point of the current path segment into position information including longitude, latitude and altitude to obtain the position information after longitude and latitude conversion; Reversely calculate the current path based on the position information, attitude matrix and odometer increment data after longitude and latitude conversion; The position information of each reverse sampling point is compared with the original data of the corresponding strapdown inertial navigation odometer integrated navigation, and the error of the system is evaluated and estimated through filtering to obtain the error estimation result; According to the result of error estimation, error compensation is performed on the navigation data of the next section, and the process of reverse deduction, estimation and compensation is repeated to achieve continuous correction and estimation of the navigation data of the mining navigation system.

2. The mining navigation solution method based on reverse deduction and forward compensation according to claim 1 is characterized in that: The location information after longitude and latitude conversion is p n =[L n λ n h n ] T ; Among them, p n Indicates the location information of the end point of the nth path; L n Indicates the latitude of the end point of the nth path; λ n Indicates the longitude of the end point of the nth path; h n represents the height of the end point of the nth path; T represents the transpose of the vector.

3. The mining navigation solution method based on reverse deduction and forward compensation according to claim 2 is characterized in that: Based on the position information, attitude matrix and odometer increment data after latitude and longitude conversion, the current path is calculated in reverse, including: Determine the end point position p of the nth path n , including latitude L n , longitude λ n and height h n , and obtain the attitude matrix, which contains the attitude information of the navigation system in three axes, namely heading, pitch and roll; Obtain odometer incremental data, including the distance and direction changes of the vehicle in each sampling time interval, and set the starting point of the reverse calculation as the end point position p n , which is the end point p of the nth path nm ; Initialize an empty list to store the position information of each sampling point calculated during the inverse process; Starting from the end point, traverse each sampling time point t in the nth segment of the path in reverse i , where i traverses from m to 1, and m represents the total number of sampling points in the nth segment; For each sampling time point t i , using the attitude matrix and odometer increment data to calculate from the current sampling point p nm To the next sampling point, t i -1 displacement vector Δp i ; For each calculated displacement vector Δp i , applied to the current position, to obtain the position coordinates p of the next sampling point i ′; When i=m-1, Among them, P' n(m-1) Represents the position coordinates of the m-1th sampling point on the nth path obtained by reverse calculation; represents the attitude matrix; D represents the distance part of the odometer increment data; R nm Represents the rotation matrix associated with the time interval from the mth sampling point to the m-1th sampling point; For each i, the position coordinates are calculated and updated using the attitude matrix, odometer increment data, and rotation matrix, and each calculated position coordinate p i 'Add to the storage list; After the reverse calculation is completed, the storage list contains the nth segment path from the starting position to the end position p nm The location information of each sampling point.

4. The mining navigation solution method based on reverse deduction and forward compensation according to claim 3 is characterized in that: The posture matrix for: Among them, cosψ and sinψ represent the cosine and sine values ​​of the yaw angle ψ; cosθ and sinθ represent the cosine and sine values ​​of the pitch angle θ; cosφ and sinφ represent the cosine and sine values ​​of the roll angle φ.

5. The mining navigation solution method based on reverse deduction and forward compensation according to claim 4 is characterized in that: The position information of each reverse sampling point is compared with the original data of the corresponding strapdown inertial navigation odometer integrated navigation. Through filtering, the system error is evaluated and estimated to obtain the error estimation results, including: Obtain the original data of the nth strapdown inertial navigation odometer integrated navigation, including the position information of each sampling time point, and obtain the position information of each sampling point; For each sampling point, the position difference between the two sets of data is calculated by Euclidean distance, and the position difference of each sampling point is formed into a position difference data set; Performing filtering processing on the position difference data set to obtain filtered position difference data; According to the position difference data after filtering, the error of the system is evaluated, including the statistical indicators of the mean, standard deviation, maximum and minimum values ​​of the error are calculated to obtain the result of the error evaluation; The error evaluation results are analyzed through the error model to obtain the error estimation results.

6. The mining navigation solution method based on reverse deduction and forward compensation according to claim 5 is characterized in that: Through the error model, the results of error assessment are analyzed to obtain the results of error estimation, including: Construct an error model and set the initial parameters of the error model; The error evaluation results, including the error mean and standard deviation of each sampling point, are input into the error model for training to obtain a trained error model; The error model after training is used to analyze the error, including the source, type, and distribution characteristics of the error, and the error estimation result is obtained, including the size, direction, and distribution information of the error.

7. The mining navigation solution method based on reverse deduction and forward compensation according to claim 6 is characterized in that: The calculation formula of Euclidean distance is: Where, ΔE i represents the weighted position difference of the i-th sampling point; w x 、w y 、w z represents the weight coefficient; Represents the coordinate values ​​of the theoretical position of the i-th measurement point in the x, y, and z directions; At time τ i When , the actual measurement position of the i-th measurement point; b x 、b y 、b z Indicates the offset in the x, y, and z directions; Represents the noise variance of the ith measurement point in the x, y, and z directions.

8. The mining navigation solution method based on reverse deduction and forward compensation according to claim 7 is characterized in that: According to the error estimation results, the next section of navigation data is compensated for errors. By repeating the process of reverse calculation, estimation and compensation, the navigation data of the mining navigation system can be continuously corrected and estimated, including: According to the error estimation result, the navigation data of the n+1th segment is compensated for the error, including adjusting the position information, correcting the attitude matrix, and calibrating the odometer to obtain the compensated navigation data; The compensated navigation data is used as input data for the next iteration, and the compensated navigation data is continuously used as a new starting point to continuously correct and estimate the navigation data.

9. A mining navigation solution system based on reverse deduction and forward compensation, the system implementing the method as claimed in any one of claims 1 to 8, characterized in that: include: Data acquisition module, used to collect navigation data in the mine in real time, including position information, speed information, direction information and environmental parameters; The reverse calculation module is used to reversely calculate the movement trajectory of the mine car based on the navigation data of the data acquisition module to obtain the previous position of the mine car; The forward compensation module is used to perform forward compensation on the reverse calculation results according to the geographical information of the mine, so as to correct the error caused by the mine environment and obtain the current position of the mine car; A map matching module is used to match the compensated mine car position with the mine map to obtain a map matching result; The navigation solution module is used to perform navigation solution based on the results of inverse calculation, forward compensation and map matching.

10. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

Images