Multi-sensor collaborative positioning system and method for underground gas pipeline positioning robot

Abstract

The present invention discloses a multi-sensor collaborative positioning system and method for an underground gas pipeline positioning robot, and relates to the field of pipeline three-dimensional positioning. The present invention installs an inertial sensor and several auxiliary sensors, such as a visual sensor, on a current positioning robot, and fuses the data measured by the inertial sensor and the several auxiliary sensors when three-dimensional positioning is performed on the gas pipeline to be positioned, thereby overcoming the problem of inaccurate positioning caused by increased measurement error of a single inertial sensor when passing through special environments such as long-distance movement, large elbows and large inclinations in the gas pipeline to be positioned; a gas pipeline environment type mapping model is constructed, so that the type of a special pipeline environment is determined and identified before the current final positioning robot enters the special pipeline environment; and a sensor error source type mapping model is constructed, so as to provide a basis for determining which performance parameter of the inertial sensor is affected in the subsequent determination.

Description

Technical Field

[0001] The present invention belongs to the field of pipeline three-dimensional positioning, and in particular, relates to a multi-sensor collaborative positioning system and method for an underground gas pipeline positioning robot. Background Art

[0002] A Chinese patent with announcement number CN106441272B discloses an automatic and rapid positioning system and method for a pipeline clearing robot. The pipeline robot emits a low-frequency electromagnetic pulse signal from the inside of the pipe, which can penetrate the pipe wall to reach the outside of the pipe for reception by a magnetoelectric sensor. The convolution sum is calculated by taking the theoretical waveform of the sensor received signal and the appropriate voltage signal, and judging the correlation between the two, and the GPS coordinate value is calculated as the pipeline robot on the xoy plane.

[0003] Robots are used to locate gas pipelines in complex environments, such as long-distance movement, large elbows, and large inclinations. Traditional positioning using a single sensor is easily affected by environmental factors such as long distances and inclinations in complex environments, thus affecting the accuracy and stability of positioning. Summary of the invention

[0004] In response to the problems in the related art, the present invention proposes a multi-sensor collaborative positioning system and method for an underground gas pipeline positioning robot to overcome the above-mentioned technical problems existing in the existing related technology.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is a multi-sensor collaborative positioning method for an underground gas pipeline positioning robot, comprising the following steps:

[0007] S1. Filter the parameter types in the current inertial sensor that are affected by the gas pipeline environment to obtain an inertial sensor parameter type set;

[0008] S2. Constructing a gas pipeline environment type mapping model using pipeline internal entrance image data corresponding to several types of gas pipeline environments; setting the parameter types measured by the gas pipeline to be located, the current positioning robot, and the corresponding auxiliary sensor to obtain an auxiliary parameter type set;

[0009] S3, obtaining data of coordinate data of several test points in the existing gas pipeline measured by the current positioning robot when the parameters in the inertial sensor parameter type set are abnormal, and obtaining a matrix set of interference measurement coordinate data of the existing test points;

[0010] S4, constructing a coordinate data image according to the existing test point interference measurement coordinate data matrix set and constructing a sensor error source type mapping model in conjunction with the inertial sensor parameter type set;

[0011] Construct multiple sensor data fusion mapping equations according to the current positioning robot, inertial sensor parameter type set and auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix;

[0012] S5, calculating the error of the corresponding data between the existing test point interference measurement coordinate data matrix set and the existing test point actual coordinate data matrix and adjusting the initial sensor data fusion mapping equation matrix to obtain the final sensor data fusion mapping equation matrix;

[0013] S6. Using the current positioning robot to perform a three-dimensional positioning operation on the gas pipeline to be positioned, in coordination with the final sensor data fusion mapping equation matrix, the gas pipeline environment type mapping model, and the sensor error source type mapping model;

[0014] Preferably, the S1 comprises the following steps:

[0015] S11, setting several underground gas pipeline environment types to obtain a gas pipeline environment type set; setting the environmental variable type to be tested according to the gas pipeline environment type set to obtain a pre-screening environmental variable type set;

[0016] S12, setting the current inertial sensor, setting several test points with known coordinate data and corresponding coordinate data, obtaining a known measurement test point set and a test point actual coordinate data set; using the current inertial sensor to collect the coordinate data of each test point in the known measurement test point set in a static environment, obtaining a test point static coordinate data set; calculating the error data between the test point actual coordinate data set and the test point static coordinate data set, obtaining static error data;

[0017] S13, in conjunction with the environmental variable type set, using the current inertial sensor to collect coordinate data of each test point in the known measurement test point set for multiple times, applying a single different type of environmental variable during each collection process and recording the corresponding coordinate data measured by the current inertial sensor, to obtain a dynamic coordinate data matrix of the test point;

[0018] Calculate the error data between each row of data in the dynamic coordinate data matrix of the test point and the actual coordinate data set of the test point to obtain a dynamic error data set;

[0019] S14, setting a test point coordinate measurement error threshold; selecting the environmental variable type corresponding to the dynamic error data whose difference between the dynamic error data set and the static error data is greater than or equal to the test point coordinate measurement error threshold, to obtain a screened environmental variable type set;

[0020] In conjunction with the filtered environmental variable type set, parameter types of a plurality of inertial sensors that interact with the filtered environmental variable types are set to obtain an inertial sensor parameter type set;

[0021] By setting the gas pipeline environment type set, it is convenient to set corresponding data fusion calculation equations for different types of gas pipeline environments, thereby providing accuracy of data fusion; by collecting and calculating static error data, and then comparing the static error data with the error data after being affected by a single different type of environmental variable, it is possible to determine which environmental variables affect the measurement accuracy of the sensor, and then based on these environmental variables, determine which performance parameters of the sensor are affected, which affects the measurement accuracy of the sensor; this provides a basis for subsequently setting corresponding data fusion equations for the way in which the performance parameters of each sensor are affected.

[0022] Preferably, S2 comprises the following steps:

[0023] S21, respectively collecting a plurality of existing and known pipeline internal inlet image data corresponding to each gas pipeline environment type in the gas pipeline environment type set to obtain a pipeline internal inlet image data matrix;

[0024] S22, constructing a gas pipeline environment type mapping model using the pipeline internal entrance image data matrix and the gas pipeline environment type set;

[0025] S23, setting the gas pipeline to be located and the current positioning robot; setting several auxiliary sensors according to the internal environmental characteristics of the gas pipeline to be located to obtain an auxiliary sensor set; installing each auxiliary sensor in the auxiliary sensor set and the current inertial sensor combination on the current positioning robot to obtain the current final positioning robot; setting the parameter type measured by each auxiliary sensor in the auxiliary sensor set and combining them to obtain an auxiliary parameter type set;

[0026] By collecting image data at the entrances of special environments in various gas pipelines, a data basis is provided for the subsequent construction of a gas pipeline environment type mapping model; a set of auxiliary parameter types is provided to provide a basis for the subsequent establishment of data fusion equations; and by constructing a gas pipeline environment type mapping model, the type of special pipeline environment can be determined and identified before the current final positioning robot enters the special pipeline environment.

[0027] Preferably, S3 comprises the following steps:

[0028] S31. For each gas pipeline environment type in the gas pipeline environment type set, a corresponding existing gas pipeline with a single environment type is set to obtain an existing gas pipeline set;

[0029] S32, evenly setting a number of test points with known coordinates on each existing gas pipeline in the existing gas pipeline set, and obtaining a data matrix of actual coordinates of the existing test points;

[0030] S33, in the process that the current final positioning robot moves multiple times inside each existing gas pipeline in the existing gas pipeline set and collects coordinate data of each test point, when the parameter corresponding to each inertial sensor parameter type in the inertial sensor parameter type set of the inertial sensor on the current final positioning robot is affected, the coordinate data of each test point is collected respectively to obtain the existing test point interference measurement coordinate data matrix set;

[0031] Providing an existing gas pipeline with a single environment type corresponding to the gas pipeline environment type set facilitates subsequent targeted testing of each type of gas pipeline environment, thereby constructing an accurate data fusion equation; obtaining the actual coordinate data matrix of the existing test points and the interference measurement coordinate data matrix set of the existing test points through measurement, provides a basis for subsequent determination whether to adjust the initial data fusion mapping equation.

[0032] Preferably, S4 comprises the following steps:

[0033] S41, constructing a coordinate data image according to each row of coordinate data in each existing test point interference measurement coordinate data matrix in the existing test point interference measurement coordinate data matrix set, to obtain an existing test point interference measurement coordinate image data matrix;

[0034] S42, constructing a sensor error source type mapping model according to the existing test point interference measurement coordinate image data matrix and the inertial sensor parameter type set;

[0035] S43, constructing a plurality of sensor data fusion mapping equations according to the current final positioning robot, the gas pipeline environment type set, the inertial sensor parameter type set, and the auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix;

[0036] By constructing a coordinate data image for each row of coordinate data in the interference measurement coordinate data matrix of each existing test point, the approximate change trend of the collected coordinate data when different types of performance parameters of the inertial sensor are affected in different types of environments is learned, which provides a data basis for the subsequent construction of the sensor error source type mapping model; by setting the sensor data fusion mapping equation under the joint action of different types of environments and different types of performance parameters of the inertial sensor, an adjustment object is provided for subsequent targeted adjustments.

[0037] Preferably, S5 comprises the following steps:

[0038] S51, calculating the Euclidean distance between each row of data in each existing test point interference measurement coordinate data matrix in the existing test point interference measurement coordinate data matrix set and the data in the corresponding row in the existing test point actual coordinate data matrix, to obtain the existing test point interference measurement coordinate error data matrix;

[0039] S52, setting an interference measurement coordinate error threshold; when there is an existing test point interference measurement coordinate error data greater than or equal to the interference measurement coordinate error threshold in the existing test point interference measurement coordinate error data matrix, the existing test point interference measurement coordinate error data and the initial sensor data fusion mapping equation in the corresponding initial sensor data fusion mapping equation matrix are respectively recorded as coordinate error data to be adjusted and initial sensor data fusion mapping equation to be adjusted; the initial sensor data fusion mapping equation to be adjusted is adjusted until there is no existing test point interference measurement coordinate error data greater than or equal to the interference measurement coordinate error threshold in the existing test point interference measurement coordinate error data matrix, and a final sensor data fusion mapping equation matrix is ​​obtained; otherwise, the initial sensor data fusion mapping equation matrix does not need to be adjusted, and the initial sensor data fusion mapping equation matrix is ​​used as the final sensor data fusion mapping equation matrix;

[0040] By calculating the Euclidean distance between each row of data in the interference measurement coordinate data matrix of each existing test point in the interference measurement coordinate data matrix set of the existing test points and the data of the corresponding row in the actual coordinate data matrix of the existing test points, the size of the measurement error caused by measuring according to the constructed initial data fusion equation can be measured; and then by setting the interference measurement coordinate error threshold, a judgment basis is provided for determining whether the current measurement error meets the requirements; and then it is determined whether the corresponding data fusion equation needs to be adjusted.

[0041] Preferably, adjusting the initial sensor data fusion mapping equation to be adjusted in S52 includes the following steps:

[0042] S521, constructing a data fusion mapping equation to adjust the African vulture population; setting the maximum number of iterations of the data fusion mapping equation to adjust the African vulture population to And the current iteration number is , respectively denoted as the maximum iteration number of equation adjustment and the current iteration number of equation adjustment;

[0043] S522, setting the coefficient corresponding to each measurement data and the value interval of each bias in the initial sensor data fusion mapping equation to be adjusted, and obtaining a measurement data coefficient value interval matrix and a bias value interval set;

[0044] According to the measurement data coefficient value interval matrix and the bias value interval set, a data fusion mapping equation is set to adjust the initial position matrix of each African vulture in the African vulture population to obtain an initial position matrix set;

[0045] S523, constructing a data fusion mapping equation to adjust the fitness function of the African vulture population according to the coordinate error data to be adjusted;

[0046] S524, start iteration, iterate and set the current iteration number of the equation adjustment to 1; in the first round of iteration, use the data fusion mapping equation to adjust the fitness function of the African vulture population to calculate the fitness value of the initial position matrix of each African vulture in the initial position matrix set, and obtain a first fitness value set; use the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the African vulture as the first global optimal fitness and the first global optimal position respectively; update the initial position matrix of each African vulture in the initial position matrix set according to the first global optimal fitness and the first global optimal position; after the update is completed, add 1 to the current iteration number of the equation adjustment and enter the next round of iteration;

[0047] In each other round of iteration, the data fusion mapping equation is used to adjust the fitness function of the African vulture population, and the fitness value of the position matrix of each African vulture in the African vulture population is calculated by using the data fusion mapping equation updated in the previous round of iteration to obtain a second fitness value set; the maximum fitness value in the second fitness value set and the corresponding position matrix of the African vulture are respectively used as the second global optimal fitness and the second global optimal position; the position matrix of each African vulture in the African vulture population is updated by the data fusion mapping equation updated in the previous round of iteration according to the second global optimal fitness and the second global optimal position; after the update is completed, the equation is adjusted to increase the current iteration number by 1 and enter the next round of iteration;

[0048] S525, when When , stop the iteration and obtain the final global optimal position; substitute each position component of the final global optimal position into the initial sensor data fusion mapping equation to be adjusted, and obtain the optimized sensor data fusion mapping equation; obtain and calculate the corresponding coordinate error data in conjunction with the optimized sensor data fusion mapping equation, and record it as the optimized coordinate error data; when the optimized coordinate error data is less than the interference measurement coordinate error threshold, the adjustment is completed; otherwise, return to S524 and continue to iterate until the optimized coordinate error data is less than the interference measurement coordinate error threshold;

[0049] In the African vulture optimization algorithm, vultures will concentrate on searching in food-rich areas based on the foraging information of other vultures during the foraging process, and can conduct more detailed searches in areas close to the optimal solution, thereby improving the accuracy of the solution. It has the advantages of few parameters and easy implementation. It can quickly converge to the vicinity of the optimal solution during the optimization process and can find a more optimized solution in a shorter time, thereby improving the efficiency of the algorithm; it has strong adaptability and robustness; based on the above advantages, the African vulture optimization algorithm is used in this scheme to iteratively adjust the coefficients and bias of the data fusion equation corresponding to the measurement error greater than or equal to the set threshold for multiple times, and the corresponding measurement error is used as the fitness function. As the iteration proceeds, the measurement error becomes smaller and smaller, and finally the measurement accuracy requirement is met.

[0050] Preferably, the S6 comprises the following steps:

[0051] S61, using the current final positioning robot to perform a three-dimensional positioning operation on the gas pipeline to be positioned; using an auxiliary sensor to obtain image data in front of the current final positioning robot in real time to obtain real-time image data; inputting the real-time image data into a gas pipeline environment type mapping model for mapping to obtain the current gas pipeline environment type; and then using an inertial sensor to collect coordinate data of the current final positioning robot for several times to obtain a real-time coordinate data set;

[0052] S62. Construct a coordinate image according to the real-time coordinate data set to obtain a current coordinate change image; input the current coordinate change image into the sensor error source type mapping model for mapping to obtain the sensor error source type; select the corresponding final sensor data fusion mapping equation in the final sensor data fusion mapping equation matrix according to the sensor error source type and the current gas pipeline environment type to fuse the data of each sensor on the current final positioning robot; until the current final positioning robot completely passes through the gas pipeline to be positioned.

[0053] The multi-sensor collaborative positioning system of the underground gas pipeline positioning robot includes a gas pipeline environment type setting module, a current inertial sensor affected parameter screening module, a gas pipeline environment type mapping model construction module, a robot positioning setting module, an existing test point actual coordinate acquisition module, an existing test point interference measurement coordinate data acquisition module, an existing test point interference measurement image construction module, a sensor error source type mapping model construction module, an initial sensor data fusion mapping equation construction module, an existing test point interference measurement coordinate error data calculation module, a sensor data fusion mapping equation adjustment module and a current three-dimensional positioning module.

[0054] The present invention has the following beneficial effects:

[0055] 1. In the present invention, an inertial sensor and several auxiliary sensors, such as a visual sensor, are installed on the current positioning robot. When the gas pipeline to be positioned is three-dimensionally positioned, the data measured by the inertial sensor and the several auxiliary sensors are fused, thereby overcoming the problem of inaccurate positioning caused by increased measurement errors of a single inertial sensor when passing through special environments such as long-distance movement, large elbows and large inclinations in the gas pipeline to be positioned.

[0056] 2. In the present invention, static error data is collected and calculated and compared with the error data after being affected by a single environmental variable of different types, so as to obtain which environmental variables affect which performance parameters of the sensor.

[0057] 3. In the present invention, a gas pipeline environment type mapping model is constructed to enable the current final positioning robot to identify the type of the special pipeline environment before entering the special pipeline environment; and a sensor error source type mapping model is constructed to provide a basis for determining which performance parameter of the inertial sensor is affected.

[0058] 4. In the present invention, by setting the sensor data fusion mapping equation under the joint action of different types of environment types and different types of performance parameters of inertial sensors, an adjustment object is provided for subsequent targeted adjustment; the African vulture optimization algorithm is used to iteratively adjust the coefficients and bias of the data fusion equation corresponding to the measurement error greater than or equal to the set threshold for multiple times. As the iteration proceeds, the measurement error becomes smaller and smaller, and finally the measurement accuracy requirement is met.

[0059] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0061] Figure 1 It is a schematic flow chart of the multi-sensor collaborative positioning method of the underground gas pipeline positioning robot of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0063] Embodiment 1

[0064] See also Figure 1 This embodiment is a multi-sensor collaborative positioning method for an underground gas pipeline positioning robot, comprising the following steps:

[0065] S1. Filter the parameter types in the current inertial sensor that are affected by the gas pipeline environment to obtain an inertial sensor parameter type set;

[0066] The S1 comprises the following steps:

[0067] S11. Set several underground gas pipeline environment types to obtain a gas pipeline environment type set. , a 1i Indicates the setting i Types of underground gas pipeline environments, Indicates the total number of underground gas pipeline environment types set; the gas pipeline environment type set includes long-distance movement, large elbows and large inclinations, etc.; according to the gas pipeline environment type set, the environmental variable type to be tested is set to obtain the environmental variable type set before screening , a 3i Indicates the setting i The types of environment variables to be tested, Indicates the total number of environmental variable types to be tested; the environmental variable types include temperature variable type, velocity variable type, acceleration variable type, distance variable type, angular velocity variable type, etc.;

[0068] S12, set the current inertial sensor, set several test points with known coordinate data and corresponding coordinate data, and obtain a known measurement test point set and a test point actual coordinate data set a 4 ; In a static environment, use the current inertial sensor to collect the coordinate data of each test point in the known measurement test point set to obtain the static coordinate data set of the test point a 5 ; a 4 , a 5 They are as follows:

[0069] ;

[0070]

[0071] in, , , Respectively represent the first i 1, 2, 3-dimensional actual coordinate data of each test point; , , They represent the current inertial sensor in a static environment to measure the known test points. i The test points are measured to obtain 1, 2, and 3-dimensional coordinate data. Indicates the total number of test points with known coordinate data;

[0072] Calculate the error data between the actual coordinate data set of the test point and the static coordinate data set of the test point to obtain the static error data The calculation formula is as follows:

[0073] ;

[0074] S13, in conjunction with the environmental variable type set, the current inertial sensor is used multiple times to collect the coordinate data of each test point in the known measurement test point set, and a single different type of environmental variable is applied during each collection process and the corresponding coordinate data measured by the current inertial sensor is recorded to obtain the dynamic coordinate data matrix of the test point a 6 ;as follows,

[0075] ;

[0076] in, , , Respectively expressed in j The first set of known measurement test points collected by the current inertial sensor under the influence of environmental variables i 1, 2, 3-dimensional coordinate data of each test point;

[0077] Calculate the error data between each row of the test point dynamic coordinate data matrix and the actual coordinate data set of the test point to obtain the dynamic error data set , Indicates the test point dynamic coordinate data matrix j The error data between the row data and the actual coordinate data set of the test point; the calculation formula is as follows,

[0078] ;

[0079] S14, setting a test point coordinate measurement error threshold; selecting the environmental variable type corresponding to the dynamic error data whose difference between the dynamic error data set and the static error data is greater than or equal to the test point coordinate measurement error threshold, and obtaining a screened environmental variable type set. , a 7i Indicates the first i Types of environment variables, Indicates the total number of selected environment variable types;

[0080] In conjunction with the filtered environmental variable type set, set the parameter types of several inertial sensors that interact with the filtered environmental variable types to obtain an inertial sensor parameter type set. , a 2i Indicates the inertial sensor to be set. i parameter types, Indicates the total number of parameter types of the set inertial sensor; the inertial sensor parameter types include accuracy index, range, etc.; the accuracy index is generally affected by temperature, and the range is generally affected by acceleration;

[0081] S2. Constructing a gas pipeline environment type mapping model using pipeline internal entrance image data corresponding to several types of gas pipeline environments; setting the parameter types measured by the gas pipeline to be located, the current positioning robot, and the corresponding auxiliary sensor to obtain an auxiliary parameter type set;

[0082] The S2 comprises the following steps:

[0083] S21, respectively collect a number of existing and known pipeline internal entrance image data corresponding to each gas pipeline environment type in the gas pipeline environment type set, and obtain a pipeline internal entrance image data matrix ;as follows,

[0084] ;

[0085] in, Indicates that for i The first type of underground gas pipeline environment collection j Existing and known pipeline internal entrance image data, Indicates that for i The total number of existing and known pipeline internal entrance image data collected for the underground gas pipeline environment type;

[0086] S22, constructing a gas pipeline environment type mapping model using the pipeline internal entrance image data matrix and the gas pipeline environment type set;

[0087] The S22 comprises the following steps:

[0088] S221, constructing a first initial CNN classification model and setting a first training data ratio; dividing the pipeline internal entrance image data matrix by row according to the first training data ratio to obtain a pipeline internal entrance image training data matrix and a pipeline internal entrance image test data matrix;

[0089] S222, setting a first training error threshold; inputting the pipeline internal entrance image training data matrix and the gas pipeline environment type set as training data and training labels into the first initial CNN classification model for training; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain a first trained CNN classification model; otherwise, continuing the training until the training error is less than the first training error threshold;

[0090] S223, setting a first test accuracy threshold; inputting the pipeline internal entrance image test data matrix and the gas pipeline environment type set as test data and test labels respectively into the first trained CNN classification model for testing; after the test is completed, obtaining the first test accuracy data; when the first test accuracy data is greater than or equal to the first test accuracy threshold, using the first trained CNN classification model as the gas pipeline environment type mapping model; otherwise, returning to S222 to continue training until the first test accuracy data is greater than or equal to the first test accuracy threshold;

[0091] S23, set the gas pipeline to be located and the current positioning robot; set several auxiliary sensors according to the internal environmental characteristics of the gas pipeline to be located to obtain an auxiliary sensor set; the type of the current positioning robot is preferably a variable diameter robot; the auxiliary sensor set includes visual sensors, etc.; each auxiliary sensor in the auxiliary sensor set and the current inertial sensor combination are installed on the current positioning robot to obtain the current final positioning robot; set the parameter type measured by each auxiliary sensor in the auxiliary sensor set and combine them to obtain an auxiliary parameter type set , d i Indicates the setting i Auxiliary parameter types, Indicates the total number of auxiliary parameter types set; auxiliary parameter types include large elbow radius and large inclination angle, etc.

[0092] S3, obtaining data of coordinate data of several test points in the existing gas pipeline measured by the current positioning robot when the parameters in the inertial sensor parameter type set are abnormal, and obtaining a matrix set of interference measurement coordinate data of the existing test points;

[0093] The S3 comprises the following steps:

[0094] S31. For each gas pipeline environment type in the gas pipeline environment type set, a corresponding existing gas pipeline with a single environment type is set to obtain an existing gas pipeline set. , b i Indicates the first i There is an existing gas pipeline with a single environment type corresponding to the gas pipeline environment type setting;

[0095] S32, uniformly setting a number of test points with known coordinates on each existing gas pipeline in the existing gas pipeline set, and obtaining a data matrix of actual coordinates of the existing test points ;as follows,

[0096] ;

[0097] in, , , Respectively represent the existing gas pipeline concentration i The first set on an existing gas pipeline j Test points with known coordinates x axis, y axis, z Axis actual coordinate data, Indicates the total number of test points with known coordinates set on each existing gas pipeline in the existing gas pipeline set;

[0098] S33, in the process that the current final positioning robot moves multiple times inside each existing gas pipeline in the existing gas pipeline set and collects coordinate data of each test point, when the parameter corresponding to each inertial sensor parameter type in the inertial sensor parameter type set of the inertial sensor on the current final positioning robot is affected, the coordinate data of each test point is collected respectively to obtain the interference measurement coordinate data matrix set of the existing test points , c k Indicates that when the inertial sensor parameter type set of the inertial sensor on the current final positioning robot is k When the parameters corresponding to the inertial sensor parameter type are affected, the existing test point interference measurement coordinate data matrix is ​​collected; as follows,

[0099] ;

[0100] in, c kijx , c kijy , c kijz Respectively c k The existing gas pipelines in China are concentrated in i The first set on an existing gas pipeline j Test points with known coordinates x axis, y axis, z Axis coordinate data;

[0101] S4, constructing a coordinate data image according to the existing test point interference measurement coordinate data matrix set and constructing a sensor error source type mapping model in conjunction with the inertial sensor parameter type set;

[0102] Construct multiple sensor data fusion mapping equations according to the current positioning robot, inertial sensor parameter type set and auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix;

[0103] The S4 comprises the following steps:

[0104] S41, constructing a coordinate data image according to each row of coordinate data in each existing test point interference measurement coordinate data matrix in the existing test point interference measurement coordinate data matrix set, and obtaining an existing test point interference measurement coordinate image data matrix ;as follows,

[0105] ;

[0106] in, Indicates the interference measurement coordinate data matrix set according to the existing test point i The existing test point interference measurement coordinate data matrix j Coordinate data image data constructed from row coordinate data;

[0107] S42, constructing a sensor error source type mapping model according to the existing test point interference measurement coordinate image data matrix and the inertial sensor parameter type set;

[0108] The S42 comprises the following steps:

[0109] S421, constructing a second initial CNN classification model and setting a second training data ratio; dividing the existing test point interference measurement coordinate image data matrix by row according to the second training data ratio to obtain the existing test point interference measurement coordinate image training data matrix and the existing test point interference measurement coordinate image test data matrix;

[0110] S222, setting a second training error threshold; inputting the existing test point interference measurement coordinate image training data matrix and the inertial sensor parameter type set as training data and training labels respectively into the second initial CNN classification model for training; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a second trained CNN classification model; otherwise, continuing the training until the training error is less than the second training error threshold;

[0111] S223, set a second test accuracy threshold; input the existing test point interference measurement coordinate image test data matrix and the inertial sensor parameter type set as test data and test labels respectively into the second trained CNN classification model for testing; after the test is completed, obtain the second test accuracy data; when the second test accuracy data is greater than or equal to the second test accuracy threshold, use the second trained CNN classification model as the sensor error source type mapping model; otherwise, return to S222 to continue training until the second test accuracy data is greater than or equal to the second test accuracy threshold;

[0112] S43: construct multiple sensor data fusion mapping equations according to the current final positioning robot, the gas pipeline environment type set, the inertial sensor parameter type set, and the auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix ;as follows,

[0113] ;

[0114] in, Indicates the first i When the parameter corresponding to the inertial sensor parameter type fails and at the same time is in the gas pipeline environment type set j The initial sensor data fusion mapping equation in the gas pipeline environment type is as follows:

[0115] ;

[0116] In the formula, , represent respectively the acceleration data and angular velocity data measured by the inertial sensor on the current final positioning robot, It indicates the first k types of auxiliary parameter data; , , Respectively Corresponding calculation of acceleration data in x axis, y axis, z Coefficients of axis coordinate data; , , Respectively Corresponding calculation of angular velocity data in x axis, y axis, z Coefficients of axis coordinate data; , , Respectively Middle k Corresponding calculations of various types of auxiliary parameter data x axis, y axis, z Coefficients of axis coordinate data; , , Respectively Medium Calculation x axis, y axis, z The offset of axis coordinate data;

[0117] S5, calculating the error of the corresponding data between the existing test point interference measurement coordinate data matrix set and the existing test point actual coordinate data matrix and adjusting the initial sensor data fusion mapping equation matrix to obtain the final sensor data fusion mapping equation matrix;

[0118] The S5 comprises the following steps:

[0119] S51, calculating the Euclidean distance between each row of data in each existing test point interference measurement coordinate data matrix in the existing test point interference measurement coordinate data matrix set and the corresponding row of data in the existing test point actual coordinate data matrix, and obtaining the existing test point interference measurement coordinate error data matrix ;as follows,

[0120] ;

[0121] in, Indicates the existing test point interference measurement coordinate data matrix set i The existing test points interfere with the measurement coordinate data matrix jThe row data and the actual coordinate data matrix of the existing test points j Euclidean distance between rows of data;

[0122] S52, setting an interference measurement coordinate error threshold; when there is an existing test point interference measurement coordinate error data in the existing test point interference measurement coordinate error data matrix that is greater than or equal to the interference measurement coordinate error threshold, the existing test point interference measurement coordinate error data and the initial sensor data fusion mapping equation in the corresponding initial sensor data fusion mapping equation matrix are respectively recorded as coordinate error data to be adjusted and an initial sensor data fusion mapping equation to be adjusted; adjusting the initial sensor data fusion mapping equation to be adjusted until no existing test point interference measurement coordinate error data greater than or equal to the interference measurement coordinate error threshold value exists in the existing test point interference measurement coordinate error data matrix, thereby obtaining a final sensor data fusion mapping equation matrix; otherwise, there is no need to adjust the initial sensor data fusion mapping equation matrix, and the initial sensor data fusion mapping equation matrix is ​​used as the final sensor data fusion mapping equation matrix;

[0123] The step of adjusting the initial sensor data fusion mapping equation to be adjusted in S52 includes the following steps:

[0124] S521. Constructing data fusion mapping equations to adjust the African vulture population , The data fusion mapping equation adjusts the first i An African vulture, e Indicates that the data fusion mapping equation adjusts the size of the African vulture population; Sets the maximum number of iterations of the data fusion mapping equation to adjust the African vulture population And the current iteration number is , respectively recorded as the maximum number of iterations of equation adjustment and the current number of iterations of equation adjustment; the search space dimension of the data fusion mapping equation to adjust the African vulture population is ;

[0125] S522, setting the coefficient corresponding to each measurement data and the value range of each bias in the initial sensor data fusion mapping equation to be adjusted, and obtaining the measurement data coefficient value range matrix And the offset value interval set , , , Respectively represent the calculation in the initial sensor data fusion mapping equation to be adjusted x axis, y axis, z The lower limit of the offset value of the axis coordinate data. , , Respectively represent the calculation in the initial sensor data fusion mapping equation to be adjusted x axis, y axis, z The upper limit of the offset value of the axis coordinate data; as follows,

[0126] ;

[0127] in, , , They respectively represent the acceleration data and angular velocity data measured by the inertial sensor on the current final positioning robot in the initial sensor data fusion mapping equation to be adjusted, and the first k The lower limit of the coefficient corresponding to the auxiliary parameter data of this type, , , They respectively represent the acceleration data, angular velocity data measured by the inertial sensor on the current final positioning robot and the first k The upper limit of the coefficient corresponding to the auxiliary parameter data of the type;

[0128] According to the measurement data coefficient value interval matrix and the offset value interval set, a data fusion mapping equation is set to adjust the initial position matrix of each African vulture in the African vulture population to obtain an initial position matrix set , The data fusion mapping equation adjusts the African vulture population j' The initial position matrix of an African vulture is as follows:

[0129] ;

[0130] in, , , , , They represent the adjustment of the data fusion mapping equation to the first j The initial position matrix of an African vulture in the initial sensor data fusion mapping equation to be adjusted is the coefficient of the acceleration data measured by the inertial sensor on the current final positioning robot, the coefficient of the angular velocity data, and the calculation x axis, y axis, z The position component of the offset dimension of the axis coordinate data; The data fusion mapping equation adjusts the African vulture populationj The initial position matrix of the African vulture is obtained by measuring the auxiliary sensor on the current final positioning robot in the initial sensor data fusion mapping equation to be adjusted. i Positional components on the coefficient dimension of the auxiliary parameter data of the type; , , , , , The calculation formulas are as follows:

[0131] ; ;

[0132] ; ;

[0133] ; ;

[0134] In the formula, row j11 , row j12 , row j13 , row j14 , row j15 , row j2i Respectively for , , , , , Generate a random number between 0 and 1;

[0135] S523, according to the coordinate error data to be adjusted Constructing data fusion mapping equation to adjust the fitness function of African vulture population f ;as follows,

[0136] ;

[0137] S524, start iteration, iterate and set the current iteration number of the equation adjustment to 1; in the first round of iteration, use the data fusion mapping equation to adjust the fitness function of the African vulture population fCalculate the fitness value of the initial position matrix of each African vulture in the initial position matrix set to obtain a first fitness value set; use the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the African vulture as the first global optimal fitness and the first global optimal position respectively; update the initial position matrix of each African vulture in the initial position matrix set according to the first global optimal fitness and the first global optimal position; after the update is completed, adjust the current iteration number of the equation by 1 and enter the next round of iteration;

[0138] In each other iteration process, the data fusion mapping equation is used to adjust the fitness function of the African vulture population. f Calculate the fitness value of the position matrix of each African vulture in the African vulture population adjusted by the data fusion mapping equation updated in the previous iteration process to obtain a second fitness value set; use the maximum fitness value in the second fitness value set and the corresponding position matrix of the African vulture as the second global optimal fitness and the second global optimal position respectively; update the position matrix of each African vulture in the African vulture population adjusted by the data fusion mapping equation updated in the previous iteration process according to the second global optimal fitness and the second global optimal position; after the update is completed, add 1 to the current iteration number of the equation adjustment and enter the next iteration;

[0139] S525, when When , stop the iteration and obtain the final global optimal position; substitute each position component of the final global optimal position into the initial sensor data fusion mapping equation to be adjusted, and obtain the optimized sensor data fusion mapping equation; obtain and calculate the corresponding coordinate error data in conjunction with the optimized sensor data fusion mapping equation, and record it as the optimized coordinate error data; when the optimized coordinate error data is less than the interference measurement coordinate error threshold, the adjustment is completed; otherwise, return to S524 and continue to iterate until the optimized coordinate error data is less than the interference measurement coordinate error threshold;

[0140] S6. Using the current positioning robot to perform a three-dimensional positioning operation on the gas pipeline to be positioned, in coordination with the final sensor data fusion mapping equation matrix, the gas pipeline environment type mapping model, and the sensor error source type mapping model;

[0141] The S6 comprises the following steps:

[0142] S61, using the current final positioning robot to perform a three-dimensional positioning operation on the gas pipeline to be positioned; using an auxiliary sensor to obtain image data in front of the current final positioning robot in real time to obtain real-time image data; inputting the real-time image data into a gas pipeline environment type mapping model for mapping to obtain the current gas pipeline environment type; and then using an inertial sensor to collect coordinate data of the current final positioning robot for several times to obtain a real-time coordinate data set;

[0143] S62. Construct a coordinate image according to the real-time coordinate data set to obtain a current coordinate change image; input the current coordinate change image into the sensor error source type mapping model for mapping to obtain the sensor error source type; select the corresponding final sensor data fusion mapping equation in the final sensor data fusion mapping equation matrix according to the sensor error source type and the current gas pipeline environment type to fuse the data of each sensor on the current final positioning robot; until the current final positioning robot completely passes through the gas pipeline to be positioned.

[0144] Embodiment 2

[0145] The present embodiment discloses a multi-sensor collaborative positioning system for an underground gas pipeline positioning robot, the system can implement the method of the above embodiment, including a gas pipeline environment type setting module, a current inertial sensor affected parameter screening module, a gas pipeline environment type mapping model construction module, a robot positioning setting module, an existing test point actual coordinate acquisition module, an existing test point interference measurement coordinate data acquisition module, an existing test point interference measurement image construction module, a sensor error source type mapping model construction module, an initial sensor data fusion mapping equation construction module, an existing test point interference measurement coordinate error data calculation module, a sensor data fusion mapping equation adjustment module and a current three-dimensional positioning module;

[0146] The gas pipeline environment type setting module is used to set several underground gas pipeline environment types to obtain a gas pipeline environment type set;

[0147] The affected parameter screening module of the current inertial sensor is used to screen the parameter types affected by the gas pipeline environment in the gas pipeline environment type set when the coordinate data of each test point in the known measurement test point set is collected by the current inertial sensor, so as to obtain an inertial sensor parameter type set;

[0148] The gas pipeline environment type mapping model construction module is used to construct a gas pipeline environment type mapping model using existing and known pipeline internal inlet image data corresponding to the gas pipeline environment type set;

[0149] The robot positioning setting module is used to set the parameter types measured by the gas pipeline to be positioned, the current final positioning robot, and the auxiliary sensor on the current final positioning robot to obtain an auxiliary parameter type set;

[0150] The existing test point actual coordinate acquisition module is used to acquire actual coordinate data of a plurality of test points in an existing gas pipeline with a single environment type corresponding to the gas pipeline environment type set, and obtain an actual coordinate data matrix of the existing test points;

[0151] The existing test point interference measurement coordinate data acquisition module is used to obtain data of the coordinate data of several test points in the existing gas pipeline measured by the current final positioning robot when the parameters in the inertial sensor parameter type set are abnormal, and obtain the existing test point interference measurement coordinate data matrix set;

[0152] The existing test point interference measurement image construction module is used to construct a coordinate data image according to each row of coordinate data in each existing test point interference measurement coordinate data matrix in the existing test point interference measurement coordinate data matrix set to obtain an existing test point interference measurement coordinate image data matrix;

[0153] The sensor error source type mapping model construction module is used to construct a sensor error source type mapping model using an existing test point interference measurement coordinate image data matrix and an inertial sensor parameter type set;

[0154] The initial sensor data fusion mapping equation construction module is used to construct multiple sensor data fusion mapping equations according to the current final positioning robot, the gas pipeline environment type set, the inertial sensor parameter type set and the auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix;

[0155] The existing test point interference measurement coordinate error data calculation module is used to obtain the existing test point interference measurement coordinate error data matrix according to the error of the corresponding data between the existing test point interference measurement coordinate data matrix set and the existing test point actual coordinate data matrix;

[0156] The sensor data fusion mapping equation adjustment module is used to adjust the initial sensor data fusion mapping equation matrix according to the existing test point interference measurement coordinate error data matrix to obtain the final sensor data fusion mapping equation matrix;

[0157] The current three-dimensional positioning module is used to cooperate with the final sensor data fusion mapping equation matrix, the gas pipeline environment type mapping model and the sensor error source type mapping model, and use the current final positioning robot to perform a three-dimensional positioning operation on the gas pipeline to be positioned.

[0158] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0159] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A multi-sensor collaborative positioning method for an underground gas pipeline positioning robot, characterized in that: The following steps are involved: S1. Filter the parameter types in the current inertial sensor that are affected by the gas pipeline environment to obtain an inertial sensor parameter type set; S2. Constructing a gas pipeline environment type mapping model using pipeline internal entrance image data corresponding to several types of gas pipeline environments; setting the parameter types measured by the gas pipeline to be located, the current positioning robot, and the corresponding auxiliary sensor to obtain an auxiliary parameter type set; S3, obtaining data of coordinate data of several test points in the existing gas pipeline measured by the current positioning robot when the parameters in the inertial sensor parameter type set are abnormal, and obtaining a matrix set of interference measurement coordinate data of the existing test points; S4, constructing a coordinate data image according to the existing test point interference measurement coordinate data matrix set and constructing a sensor error source type mapping model in conjunction with the inertial sensor parameter type set; Construct multiple sensor data fusion mapping equations according to the current positioning robot, inertial sensor parameter type set and auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix; S5, calculating the error of the corresponding data between the existing test point interference measurement coordinate data matrix set and the existing test point actual coordinate data matrix and adjusting the initial sensor data fusion mapping equation matrix to obtain the final sensor data fusion mapping equation matrix; S6. Cooperating with the final sensor data fusion mapping equation matrix, the gas pipeline environment type mapping model and the sensor error source type mapping model, the current positioning robot is used to perform three-dimensional positioning operations on the gas pipeline to be positioned.

2. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 1 is characterized in that: The S1 comprises the following steps: S11. Setting several underground gas pipeline environment types to obtain a gas pipeline environment type set; S12, setting the current inertial sensor, setting several test points with known coordinate data and corresponding coordinate data, obtaining a known measurement test point set and a test point actual coordinate data set; using the current inertial sensor to collect the coordinate data of each test point in the known measurement test point set in a static environment and calculating the error data between the coordinate data and the test point actual coordinate data set, obtaining static error data; S13, under the condition of applying a single different type of environmental variable, using the current inertial sensor to collect the coordinate data of each test point in the known measurement test point set for multiple times, to obtain a dynamic coordinate data matrix of the test point; respectively calculating the error data between each row of data in the dynamic coordinate data matrix of the test point and the actual coordinate data set of the test point, to obtain a dynamic error data set; S14, setting a test point coordinate measurement error threshold; selecting the environmental variable type corresponding to the dynamic error data whose difference between the dynamic error data set and the static error data is greater than or equal to the test point coordinate measurement error threshold, to obtain a screened environmental variable type set; and setting parameter types of several inertial sensors that interact with the screened environmental variable type set, to obtain an inertial sensor parameter type set.

3. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 2 is characterized in that: The S2 comprises the following steps: S21, respectively collecting a plurality of existing and known pipeline internal inlet image data corresponding to each gas pipeline environment type in the gas pipeline environment type set to obtain a pipeline internal inlet image data matrix; S22, constructing a gas pipeline environment type mapping model using the pipeline internal entrance image data matrix and the gas pipeline environment type set; S23. According to the internal environmental characteristics of the gas pipeline to be located, several auxiliary sensors are set to obtain an auxiliary sensor set; each auxiliary sensor in the auxiliary sensor set and the current inertial sensor combination are installed on the current positioning robot to obtain the current final positioning robot; the parameter type measured by each auxiliary sensor in the auxiliary sensor set is set and combined to obtain an auxiliary parameter type set.

4. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 3 is characterized in that: The S3 comprises the following steps: S31. For each gas pipeline environment type in the gas pipeline environment type set, a corresponding existing gas pipeline with a single environment type is set to obtain an existing gas pipeline set; S32, evenly setting a number of test points with known coordinates on each existing gas pipeline in the existing gas pipeline set, and obtaining a data matrix of actual coordinates of the existing test points; S33, each inertial sensor parameter of the inertial sensor on the current final positioning robot is affected respectively and the coordinate data of each test point is collected to obtain an interference measurement coordinate data matrix set of existing test points.

5. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 4 is characterized in that: The S4 comprises the following steps: S41, constructing a coordinate data image according to the existing test point interference measurement coordinate data matrix set to obtain an existing test point interference measurement coordinate image data matrix; S42, constructing a sensor error source type mapping model according to the existing test point interference measurement coordinate image data matrix and the inertial sensor parameter type set; S43, constructing a plurality of sensor data fusion mapping equations according to the current final positioning robot, the gas pipeline environment type set, the inertial sensor parameter type set and the auxiliary parameter type set to obtain an initial sensor data fusion mapping equation matrix.

6. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 5 is characterized in that: The S5 comprises the following steps: S51, calculating the Euclidean distance between each row of data in the existing test point interference measurement coordinate data matrix set and the corresponding row of data in the existing test point actual coordinate data matrix, to obtain the existing test point interference measurement coordinate error data matrix; S52, setting an interference measurement coordinate error threshold; when the error between each row of data in the existing test point interference measurement coordinate data matrix set and the corresponding actual coordinate data is greater than or equal to the interference measurement coordinate error threshold, adjusting the initial sensor data fusion mapping equation corresponding to the existing test point interference measurement coordinate error data to obtain a final sensor data fusion mapping equation matrix; otherwise, the initial sensor data fusion mapping equation matrix is ​​used as the final sensor data fusion mapping equation matrix.

7. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 6 is characterized in that: In S52, adjusting the initial sensor data fusion mapping equation corresponding to the existing test point interference measurement coordinate error data includes the following steps: S521, recording the initial sensor data fusion mapping equation corresponding to the existing test point interference measurement coordinate error data described in S52 as the initial sensor data fusion mapping equation to be adjusted; constructing a data fusion mapping equation to adjust the African vulture population; setting the maximum number of iterations of the data fusion mapping equation to adjust the African vulture population is And the current number of iterations is , respectively denoted as the maximum iteration number of equation adjustment and the current iteration number of equation adjustment; S522, setting a data fusion mapping equation to adjust the initial position matrix of each African vulture in the African vulture population to obtain an initial position matrix set; S523, constructing a data fusion mapping equation to adjust the fitness function of the African vulture population; S524, start iteration; in each round of iteration, use the data fusion mapping equation to adjust the fitness function of the African vulture population, calculate the fitness value of the position matrix of each African vulture in the African vulture population adjusted by the data fusion mapping equation updated in the previous round of iteration, and update the position matrix of each African vulture in the African vulture population adjusted by the data fusion mapping equation updated in the previous round of iteration; S525, when When , stop the iteration and obtain the final global optimal position; substitute the final global optimal position into the initial sensor data fusion mapping equation to be adjusted to obtain the optimized sensor data fusion mapping equation; obtain and calculate the corresponding coordinate error data in conjunction with the optimized sensor data fusion mapping equation, which is recorded as the optimized coordinate error data; when the optimized coordinate error data is less than the interference measurement coordinate error threshold, the adjustment is completed; otherwise, return to S524 to continue iterating until the optimized coordinate error data is less than the interference measurement coordinate error threshold.

8. The multi-sensor collaborative positioning method of the underground gas pipeline positioning robot according to claim 7, characterized in that: The S6 comprises the following steps: S61, using an auxiliary sensor to obtain image data in front of the current final positioning robot in real time to obtain real-time image data; inputting the real-time image data into a gas pipeline environment type mapping model for mapping to obtain the current gas pipeline environment type; and then using an inertial sensor to collect coordinate data of the current final positioning robot for several times to obtain a real-time coordinate data set; S62. Construct a coordinate image according to the real-time coordinate data set and input it into the sensor error source type mapping model for mapping to obtain the sensor error source type; select the corresponding final sensor data fusion mapping equation in the final sensor data fusion mapping equation matrix according to the sensor error source type and the current gas pipeline environment type to fuse the data of each sensor on the current final positioning robot; until the current final positioning robot completely passes through the gas pipeline to be positioned.

9. A system for implementing the multi-sensor collaborative positioning method of the underground gas pipeline positioning robot as described in any one of claims 1 to 8.

Images