CN118500392B - Underwater robot DVL speed measurement error correction method based on improved ELM - Google Patents
Underwater robot DVL speed measurement error correction method based on improved ELM Download PDFInfo
- Publication number
- CN118500392B CN118500392B CN202410971782.7A CN202410971782A CN118500392B CN 118500392 B CN118500392 B CN 118500392B CN 202410971782 A CN202410971782 A CN 202410971782A CN 118500392 B CN118500392 B CN 118500392B
- Authority
- CN
- China
- Prior art keywords
- dvl
- output
- elm
- data
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000005259 measurement Methods 0.000 title claims abstract description 36
- 238000012937 correction Methods 0.000 title abstract description 12
- 238000012549 training Methods 0.000 claims abstract description 41
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 230000033001 locomotion Effects 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 36
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 230000008030 elimination Effects 0.000 claims description 5
- 238000003379 elimination reaction Methods 0.000 claims description 5
- 210000004027 cell Anatomy 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 230000006798 recombination Effects 0.000 claims description 3
- 238000005215 recombination Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims 1
- 238000009434 installation Methods 0.000 abstract description 3
- 230000007423 decrease Effects 0.000 abstract 1
- 230000002159 abnormal effect Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/203—Specially adapted for sailing ships
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C25/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C25/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
- G01C25/005—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass initial alignment, calibration or starting-up of inertial devices
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Manufacturing & Machinery (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses an underwater robot Doppler log (DVL) speed measurement error correction method based on an improved Extreme Learning Machine (ELM). The purpose is to improve the accuracy and reliability of the SINS/DVL navigation system of the underwater robot. The invention applies the improved ELM model to a DVL speed error correction method, and the whole process comprises the following steps: the method comprises the steps of collecting output DVL measurement data of DVL equipment of underwater robot motion, preprocessing, generating a training set and a testing set of a DVL speed correction model, constructing a DVL speed prediction model of an improved ELM mixing method, training an ELM model under the condition that GPS signals are effective, and outputting the DVL speed prediction. The invention can solve the problem of the decline of the underwater navigation precision of the underwater robot caused by inconsistent installation angles of the inertial measurement unit and the DVL in the underwater robot navigation system.
Description
Technical Field
The invention belongs to the technical field of underwater robot navigation and positioning, and particularly relates to an underwater robot DVL speed measurement error correction method based on improved ELM.
Background
Ocean resources are abundant, and ocean resources are developed with great potential, but in view of the specificity and complexity of ocean environments, exploration and development thereof face a series of challenges. Traditional manual exploration and development of ocean resources is costly and presents a great hazard. Therefore, autonomous underwater robots have been developed which can help us to develop ocean resources by controlling the underwater robots. Because they can operate in deep sea environments and can be detected and monitored for long periods of time, not only can costs be significantly reduced, but also the risk of personnel exposure to hazardous environments can be reduced.
Navigation positioning of underwater robots is critical to their safety and task performance. Inertial navigation and DVL (Doppler Velocity Log) are common underwater navigation positioning techniques, but their accuracy is affected by the installation error angle. These errors may result from inaccurate installation of the equipment or changes in the underwater environment, such as underwater terrain, currents, etc. For the impact of environmental changes on the mounting angle estimation technique, there is also a need to consider increasing the adaptability and robustness to environmental changes when designing navigation systems. This may involve the study of dynamic calibration methods, as well as the real-time processing and adjustment of sensor data to accommodate different underwater environmental conditions.
Disclosure of Invention
Aiming at the problems, the invention provides an improved ELM-based underwater robot DVL speed measurement error correction method, firstly, abnormal value detection and rejection substitution are carried out on DVL speed information in an acquired underwater robot navigation system, and the invention provides an abnormal value detection and rejection method based on least square method trend item modeling and a Schottky rule; secondly, generating a training set of a DVL speed correction model by using the speed information, constructing a DVL speed prediction model based on an improved ELM based on a mixed weighted activation function, and training the model under the condition that GPS signals are effective; finally, a DVL speed correction model is built based on the method, and an accurate speed information output function is completed.
The above purpose is achieved by the following technical scheme:
An underwater robot DVL speed measurement error correction method based on improved ELM comprises the following steps:
Step 1, acquiring output DVL measurement data of DVL equipment of underwater robot motion for preprocessing;
Step 2, the preprocessed DVL measurement data in the step 1 is formed into a sample set Three-dimensional speed information of synchronous acquisition robot under GPS signal effective condition motion carrier coordinate system constitutes sample setSample setAnd a sample setRespectively combining the data of the DVL speed prediction model training stage and the data of the model testing stage according to time segments;
step 3, constructing a DVL speed prediction model for improving the ELM mixing method;
Step 4, training the DVL speed prediction model of the improved ELM mixing method constructed in the step 3 under the condition that GPS signals are effective;
And 5, synchronously acquiring speed information of inertial navigation, DVL and GPS, inputting the test set in the step 2 into the DVL speed prediction model of the improved ELM mixing method trained in the step 4, outputting error-compensated DVL speed information by the model, inputting the error-compensated DVL speed information into an inertial navigation and DVL combined navigation system, and simultaneously comparing the positioning error of the model with the positioning error of the inertial navigation and GPS combined navigation system and the positioning error of the inertial navigation and DVL combined navigation system under original DVL data, and verifying the accuracy of the ELM model.
Further, the preprocessing of the data in step 1 specifically includes:
the sampling frequency of the output of DVL equipment is firstly based on least square method trend item modeling and the Showy-Fresnel rule Is a discrete DVL volume data sequenceWhereinU is the data length, outlier rejection processing is carried out, and K-order fitting polynomials are constructed according to trend characteristics of DVL measurement data：
，
Wherein, Coefficients for a fitting polynomial;
introducing a residual sum-of-squares function ：
，
And sum of squares function of residual errorsTake minima and pass through the coefficients of the fitting polynomialObtaining the offset guide to obtain zero
，
Calculating and expanding recombination to obtain:
，
wherein, Variables used to index the different orders of the polynomial, respectively;
obtaining fitting polynomial coefficients by matrix solution And a trend term fitting polynomial is obtained, so that the trend term of the DVL measurement data is removedThe method comprises the following steps:
，
residual sequence of DVL measurement data The method comprises the following steps:
，
wherein, Removing a mean value of a calculation formula of a trend term for DVL measurement data, wherein the calculation formula is as follows:
，
defining absolute values of residual sequences Satisfy the following requirementsIs suspicious data, i.e. wild value of DVL output, whereinThe coefficient fitting formula is as follows:； The standard deviation of the sequence after removing the trend term for DVL data is calculated as follows: ；
If the point data is determined to be the wild value, the point data is subjected to elimination processing and replaced by the mean value of the calculation formula of the DVL measurement data trending term And eliminating the outlier in the DVL data output so far, namely finishing the preprocessing of the DVL measurement data.
Further, the training set of the DVL speed prediction model in step 2 includes:
2a. Step1 the preprocessed DVL measurement data form a sample set WhereinRespectively the velocity information of the DVL in the x, y and z directions under the carrier coordinate system;
2b, synchronously acquiring three-dimensional speed information of the robot under the carrier coordinate system under the condition that GPS signals are effective to form a sample set WhereinThe speed information of the robot in the x, y and z directions under the carrier coordinate system is respectively obtained;
2c, will 、The data of the two sample sets are combined into a training set of a DVL speed prediction model training stage and a test set of a model test stage.
Further, the construction of the DVL velocity prediction model of the improved ELM hybrid method described in step 3 includes the following sub-steps:
3a, constructing a single hidden layer feedforward algorithm model which comprises an input layer, a hidden layer and an output layer 3 part; wherein the input layer comprises 3 data channels, the hidden layer comprises 12 data channels, and the output layer comprises 3 data channels;
mixing weighted activation functions and determining an optimal ELM activation function in later experiments ：
，
Wherein the method comprises the steps ofAndRespectively an implicit layer parameter weight and a bias vector, wherein V is input;
3c, constructing an ELM model output: ，
wherein, To and ideal outputCorresponding actual output; To conceal the weight matrix between the layers and the output layer, ；Representing the weight vector of the i-th cell and the hidden layer relative to each other,; The superscript T denotes the transpose of the matrix,Representation ofAnd (3) withIs an inner product of (2); balancing the trade-off relation between the original objective function and the norm penalty term for the regularization parameter; Is the weight of the input matrix and, Is the norm of the weights of the input matrix; as a result of the offset vector, L is the number of hidden layers, m is the dimension of the output vector;
3d. Simplifying the output in step 3c to ，
Wherein, Is the output matrix of the model, equivalent，；For the output weight matrix of the hidden layer,，，Is an input vector;
Calculating the hidden layer output weight matrix And weights between hidden layer and output layerI.e. solvingLeast-squares solution of (2)：Obtained by orthogonalization method, when matrixWhen the value is non-singular, then。
Further, in step 4, the ELM model is trained under the condition that the GPS signal is valid, and the specific method is as follows:
4a, inputting the training set in the step 2 into a DVL speed prediction model of the improved ELM mixing method constructed in the step 3 for training;
4b, randomly setting the weight of the input matrix according to the continuous probability distribution And offset vector；
4C, calculating the hidden layer output weight matrixAnd weights between hidden layer and output layerI.e. solvingLeast-squares solution of (2)：Obtained by orthogonalization method, when matrixWhen the value is non-singular, then；
4D, selecting the root mean square error RMSE as a judgment standard of whether training is finished,，Is an ideal output;
And 4e, judging whether the training samples are enough if the model errors do not meet the RMSE requirement, returning to the step 1 if the training samples are insufficient, otherwise, increasing the number of neurons in the hidden layer, returning to the step 3, and then training the model.
Compared with the prior art, the invention has the beneficial effects that:
(1) Compared with the prior art, the invention provides the abnormal value detection and elimination method based on the least square method trend item modeling and the Showway criterion, so that the accuracy and reliability of DVL speed information are further improved;
(2) Compared with the prior art, the invention adopts an Extreme Learning Machine (ELM) to improve the accuracy of DVL speed information, and improves the existing ELM model based on the mixed weighted activation function, thereby further improving the efficiency and accuracy of DVL speed correction.
Drawings
FIG. 1 is a flow chart of an improved ELM-based underwater robot DVL speed error correction method of the present invention;
FIG. 2 is a graph of velocity predictions for three directions of a DVL using the method of the present invention;
FIG. 3 is a graph of DVL speed error corrected using the method of the present invention.
Detailed Description
As shown in fig. 1, the method for correcting the speed measurement error of the underwater robot DVL based on the improved ELM of the present invention comprises the following steps:
Step1, data preprocessing:
In order to prevent the effect of the whole model from being influenced by abnormal values of the DVL during the movement of the underwater robot, abnormal points in the output DVL measurement data of the DVL equipment for acquiring the movement of the underwater robot are removed and replaced before model construction training is carried out, so that the accuracy of the model is ensured.
Output DVL measurement data of DVL equipment for acquiring motion of underwater robot is preprocessed, and sampling frequency is set to be based on least square trend item modeling and Showway criteriaIs a discrete DVL data sequence of (2)Performing outlier rejection processing, (wherein U is the data length). Firstly, constructing a K-order fitting polynomial according to trend characteristics of DVL data：
，
Wherein, Coefficients for a fitting polynomial;
introducing a residual sum-of-squares function ：
，
And sum of squares function of residual errorsTake minima and pass through the coefficients of the fitting polynomialObtaining the offset guide to obtain zero
，
Calculating and expanding recombination to obtain:
，
wherein, Variables used to index the different orders of the polynomial, respectively;
obtaining fitting polynomial coefficients by matrix solution And a trend term fitting polynomial is obtained, so that the trend term of the DVL measurement data is removedThe method comprises the following steps:
，
residual sequence of DVL measurement data The method comprises the following steps:
，
wherein, Removing a mean value of a calculation formula of a trend term for DVL measurement data, wherein the calculation formula is as follows:
，
defining absolute values of residual sequences Satisfy the following requirementsIs suspicious data, i.e. wild value of DVL output, whereinThe coefficient fitting formula is as follows:； The standard deviation of the sequence after removing the trend term for DVL data is calculated as follows: ；
If the point data is determined to be the wild value, the point data is subjected to elimination processing and replaced by the mean value of the calculation formula of the DVL measurement data trending term And eliminating the outlier in the DVL data output so far, namely finishing the preprocessing of the DVL measurement data, and preparing for the construction of a subsequent DVL speed training model.
Step 2, generating a training set of a DVL speed prediction model training stage and a testing set of a model testing stage:
2a. Step1 the preprocessed DVL measurement data form a sample set WhereinRespectively the velocity information of the DVL in the x, y and z directions under the carrier coordinate system;
2b, synchronously acquiring three-dimensional speed information of the robot under the carrier coordinate system under the condition that GPS signals are effective to form a sample set WhereinThe speed information of the robot in the x, y and z directions under the carrier coordinate system is respectively obtained;
2c, will 、The data of the two sample sets are combined into a training set of a DVL speed prediction model training stage and a test set of a model test stage.
Step 3, constructing a DVL speed prediction model for improving the ELM mixing method:
3a, constructing a single hidden layer feedforward algorithm model which comprises an input layer, a hidden layer and an output layer 3 part; wherein the input layer comprises 3 data channels, the hidden layer comprises 12 data channels, and the output layer comprises 3 data channels;
mixing weighted activation functions and determining an optimal ELM activation function in later experiments ：
，
Wherein the method comprises the steps ofAndRespectively an implicit layer parameter weight and a bias vector, wherein V is input;
3c, constructing an ELM model output: ，
wherein, To and ideal outputCorresponding actual output; To conceal the weight matrix between the layers and the output layer, ；Representing the weight vector of the i-th cell and the hidden layer relative to each other,; The superscript T denotes the transpose of the matrix,Representation ofAnd (3) withIs an inner product of (2); balancing the trade-off relation between the original objective function and the norm penalty term for the regularization parameter; Is the weight of the input matrix and, Is the norm of the weights of the input matrix; as a result of the offset vector, L is the number of hidden layers, m is the dimension of the output vector;
3d. Simplifying the output in step 3c to ，
Wherein, Is the output matrix of the model, equivalent，；For the output weight matrix of the hidden layer,，，Is an input vector;
Calculating the hidden layer output weight matrix And weights between hidden layer and output layerI.e. solvingLeast-squares solution of (2)：Obtained by orthogonalization method, when matrixWhen the value is non-singular, then。
Step 4, training an ELM model under the condition that GPS signals are effective:
4a, inputting the training set in the step 2 into a DVL speed prediction model of the improved ELM mixing method constructed in the step 3 for training;
4b, randomly setting the weight of the input matrix according to the continuous probability distribution And offset vector；
4C, calculating the hidden layer output weight matrixAnd weights between hidden layer and output layerI.e. solvingLeast-squares solution of (2)：Obtained by orthogonalization method, when matrixWhen the value is non-singular, then；
4D, selecting the root mean square error RMSE as a judgment standard of whether training is finished,，Is an ideal output;
And 4e, judging whether the training samples are enough if the model errors do not meet the RMSE requirement, returning to the step 1 if the training samples are insufficient, otherwise, increasing the number of neurons in the hidden layer, returning to the step 3, and then training the model.
Step5, testing a DVL speed error model:
synchronously acquiring speed information of inertial navigation, DVL and GPS, inputting the test set in the step 2 into a trained ELM model, outputting error-compensated DVL speed information by the model, inputting the error-compensated DVL speed information into an inertial navigation and DVL integrated navigation system, and simultaneously comparing the positioning error of the model with the positioning error of the inertial navigation and GPS integrated navigation system, the positioning error of the inertial navigation and DVL integrated navigation system under original DVL data, and verifying the accuracy of the ELM model.
And (3) experimental verification:
In order to verify the effectiveness of the method, experimental verification is designed. The experimental facility includes: GPS devices, DVL devices, navigation computers, IMU devices, etc. The navigation computer is responsible for collecting GPS, DVL and IMU information; the algorithm provided by the invention runs in the navigation computer. FIG. 2 shows velocity prediction graphs for three directions of DVL using the method of the invention; fig. 3 shows a graph of the DVL velocity error corrected using the method of the present invention. It can be seen from fig. 2 that the invention can predict the output of the DVL signal in case of failure, and from fig. 3, the accuracy of the predicted speed is relatively high, so that the DVL speed error can be well corrected.
Claims (1)
1. The method for correcting the DVL speed measurement error of the underwater robot based on the improved ELM is characterized by comprising the following steps:
Step 1, acquiring output DVL measurement data of DVL equipment of underwater robot motion for preprocessing;
Step 2, the preprocessed DVL measurement data in the step 1 is formed into a sample set V DVL, three-dimensional speed information of the robot under a motion carrier coordinate system under the condition that GPS signals are effective is synchronously collected to form a sample set V GPS, and data of the sample set V DVL and the sample set V GPS are respectively combined into a training set of a DVL speed prediction model training stage and a testing set of a model testing stage according to time segments;
step 3, constructing a DVL speed prediction model for improving the ELM mixing method;
Step 4, training the DVL speed prediction model of the improved ELM mixing method constructed in the step 3 under the condition that GPS signals are effective;
Step 5, synchronously acquiring speed information of inertial navigation, DVL and GPS, inputting the test set in the step 2 into the DVL speed prediction model of the improved ELM mixing method trained in the step 4, outputting error-compensated DVL speed information by the DVL speed prediction model of the improved ELM mixing method, inputting the error-compensated DVL speed information into the inertial navigation and DVL integrated navigation system, and simultaneously comparing the positioning error of the DVL speed prediction model of the improved ELM mixing method with the positioning error of the inertial navigation and GPS integrated navigation system and the positioning error of the inertial navigation and DVL integrated navigation system under original DVL data, and verifying the accuracy of the DVL speed prediction model of the improved ELM mixing method;
the pretreatment in the step 1 specifically comprises the following steps:
Firstly, based on least square method trend item modeling and a Showy Fresnel criterion, carrying out outlier elimination processing on a discrete DVL volume data sequence V DVL (U) with the sampling frequency of f D of the output of the DVL device, wherein u=1, 2,3, …, U and U are data lengths, and constructing a K-order fitting polynomial g (U) according to trend characteristics of DVL measurement data:
where u=1, 2,3, … U, K =0, 1,2, …, K, b k are coefficients of the fitting polynomial;
A residual square sum function R (u) is introduced:
And taking minimum value for residual square sum function R (u), and taking partial derivative for zero by coefficient b k of fitting polynomial to obtain
Calculating and expanding recombination to obtain:
Wherein r=0, 1,2, …, K, r, K are variables used to index the different orders of the polynomial, respectively;
obtaining fitting polynomial coefficients b k and trend term fitting polynomials by means of matrix solution, thus removing trend terms of DVL measurement data The method comprises the following steps:
wherein u=1, 2,3, …, U;
the residual sequence c DVL (u) of DVL measurement data is:
wherein, Removing a mean value of a calculation formula of a trend term for DVL measurement data, wherein the calculation formula is as follows:
The point data with absolute value |c DVL (u) | meeting the requirement of |c DVL (u) | > adelta is defined as suspicious data, namely, the wild value output by DVL, wherein a is a Showy Fresnel criterion coefficient, and a coefficient fitting formula is as follows: a= 0.2688ln (U-20) +1.63; delta is the standard deviation of the sequence after DVL data trend removal, and the calculation formula is as follows:
If the point data is determined to be the wild value, the point data is subjected to elimination processing and replaced by the mean value of the calculation formula of the DVL measurement data trending term Eliminating the outlier in the DVL data output so as to obtain the preprocessing of the DVL measurement data;
the training set of the DVL speed prediction model in step 2 includes:
2a. Step1 the preprocessed DVL measurement data form a sample set Wherein the method comprises the steps ofRespectively the velocity information of the DVL in the x, y and z directions under the carrier coordinate system;
2b, synchronously acquiring three-dimensional speed information of the robot under the carrier coordinate system under the condition that GPS signals are effective to form a sample set Wherein the method comprises the steps ofThe speed information of the robot in the x, y and z directions under the carrier coordinate system is respectively obtained;
2c, combining the data of the two sample sets V DVL、VGPS into a training set of a DVL speed prediction model training stage and a testing set of a model testing stage;
The step3 of constructing a DVL speed prediction model for improving the ELM mixing method comprises the following substeps:
3a, constructing a single hidden layer feedforward algorithm model which comprises an input layer, a hidden layer and an output layer 3 part; wherein the input layer comprises 3 data channels, the hidden layer comprises 12 data channels, and the output layer comprises 3 data channels;
Mixing the weighted activation functions and determining the optimal ELM activation function g (ω, V, b) in later experiments:
Wherein ω and b are respectively an implicit layer parameter weight and a bias vector, and V is an input;
3c, constructing an output of a DVL speed prediction model for improving the ELM mixing method:
wherein F (V j) is the actual output corresponding to the ideal output V GPSj; beta is the weight matrix between the hidden layer and the output layer, W i represents the weight vector between the ith cell and the hidden layer, w i＝(wi1,wi2,…,wiL)T; the superscript T represents the transposition of the matrix, and the xi is the trade-off relation between the regularization parameter balance original objective function and the norm penalty term; w is the weight of the input matrix and ii W 2 is the norm of the weight of the input matrix; b is a bias vector, b= (b 1,b2,...bi,...bL)T, L is the number of hidden layers, m is the dimension of the output vector;
Simplifying the output in step 3c to F (V j) =s=hβ, where S is the output matrix of the DVL velocity prediction model that improves the ELM mixing method, equivalent to F (V j), H is an output weight matrix of the hidden layer, h=h (W, b) = (H ij)U×L,hij＝g(wi,xj,bi),xj is an input vector;
Calculating a hidden layer output weight matrix H and weights beta between the hidden layer and the output layer, namely solving a minimum norm least square solution of S=H2 When the matrix HH T is a non-singular value, as determined by orthogonalization
And step4, training a DVL speed prediction model of an improved ELM mixing method under the condition that GPS signals are effective, wherein the specific method is as follows:
4a, inputting the training set in the step 2 into a DVL speed prediction model of the improved ELM mixing method constructed in the step 3 for training;
4b, randomly setting the weight W and the bias vector b of the input matrix according to the continuous probability distribution;
4c, calculating the hidden layer output weight matrix H and the weight beta between the hidden layer and the output layer, namely solving the minimum norm least square solution of T j =H2 When the matrix HH T is a non-singular value, as determined by orthogonalization
4D, selecting the root mean square error RMSE as a judgment standard of whether training is finished,
And 4e, if the model error does not meet the RMSE requirement, judging whether the training sample is enough, if the training sample is insufficient, returning to the step 1, otherwise, increasing the number of neurons of the hidden layer, returning to the step 3, and then training a DVL speed prediction model of the improved ELM mixing method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410971782.7A CN118500392B (en) | 2024-07-19 | 2024-07-19 | Underwater robot DVL speed measurement error correction method based on improved ELM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410971782.7A CN118500392B (en) | 2024-07-19 | 2024-07-19 | Underwater robot DVL speed measurement error correction method based on improved ELM |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118500392A CN118500392A (en) | 2024-08-16 |
| CN118500392B true CN118500392B (en) | 2024-09-24 |
Family
ID=92236985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410971782.7A Active CN118500392B (en) | 2024-07-19 | 2024-07-19 | Underwater robot DVL speed measurement error correction method based on improved ELM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118500392B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242194A (en) * | 2018-09-25 | 2019-01-18 | 东北大学 | A kind of thickener underflow concentration prediction method based on mixed model |
| CN109634308A (en) * | 2019-01-16 | 2019-04-16 | 中国海洋大学 | Based on intelligent navigation method under dynamic (dynamical) rate pattern auxiliary water |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2334199C1 (en) * | 2007-03-19 | 2008-09-20 | Закрытое акционерное общество "Лазекс" | Inertial-satellite navigation system with combination application of satellite data |
| CN113847915B (en) * | 2021-09-24 | 2023-12-19 | 中国人民解放军战略支援部队信息工程大学 | Navigation method of strapdown inertial navigation/Doppler integrated navigation system |
| CN114440878B (en) * | 2022-01-27 | 2023-08-15 | 湖南大学无锡智能控制研究院 | SINS and DVL integrated navigation method, device and system |
| CN116338655A (en) * | 2023-02-22 | 2023-06-27 | 河海大学 | DVL speed error calibration method based on DMD-LSTM model |
| CN117994584A (en) * | 2024-02-07 | 2024-05-07 | 北京理工大学 | Extreme learning machine image classification method based on OLQR algorithm training |
-
2024
- 2024-07-19 CN CN202410971782.7A patent/CN118500392B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242194A (en) * | 2018-09-25 | 2019-01-18 | 东北大学 | A kind of thickener underflow concentration prediction method based on mixed model |
| CN109634308A (en) * | 2019-01-16 | 2019-04-16 | 中国海洋大学 | Based on intelligent navigation method under dynamic (dynamical) rate pattern auxiliary water |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118500392A (en) | 2024-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111798491B (en) | Maneuvering target tracking method based on Elman neural network | |
| CN111985093A (en) | Adaptive unscented Kalman filtering state estimation method with noise estimator | |
| CN113191087B (en) | Navigation type depth measurement data profile sound velocity correction method combining historical profile sound velocity and actually measured surface layer sound velocity | |
| CN107451392B (en) | method for predicting residual life containing multiple correlated degradation processes | |
| CN110850371B (en) | High-resolution sound source positioning method based on Green function correction | |
| CN108489498A (en) | A kind of AUV collaborative navigation methods without mark particle filter based on maximum cross-correlation entropy | |
| Kaloop et al. | Multi input–single output models identification of tower bridge movements using GPS monitoring system | |
| CN101221238A (en) | Dynamic deviation estimation method based on gauss average value mobile registration | |
| CN103940433A (en) | Satellite attitude determining method based on improved self-adaptive square root UKF (Unscented Kalman Filter) algorithm | |
| CN107633256A (en) | Joint objective positioning and sensor registration method under a kind of multi-source ranging | |
| CN107688554A (en) | Frequency domain identification method based on adaptive Fourier decomposition | |
| Han et al. | Quadratic-Kalman-filter-based sensor fault detection approach for unmanned aerial vehicles | |
| Xie et al. | Estimating the probability density function of remaining useful life for wiener degradation process with uncertain parameters | |
| CN114925615A (en) | Radioactive leakage source positioning method and system | |
| CN115451803B (en) | Water gate real-time monitoring and early warning method based on fusion of GNSS and measuring robot | |
| CN111551897B (en) | TDOA (time difference of arrival) positioning method based on weighted multidimensional scaling and polynomial root finding under sensor position error | |
| Jwo et al. | Critical remarks on the linearised and extended Kalman filters with geodetic navigation examples | |
| CN118500392B (en) | Underwater robot DVL speed measurement error correction method based on improved ELM | |
| CN113204909B (en) | Satellite geometric feature and attitude estimation method based on ground observation photometric signal | |
| Liu et al. | LGC-Net: A lightweight gyroscope calibration network for efficient attitude estimation | |
| CN111553954B (en) | Online luminosity calibration method based on direct method monocular SLAM | |
| CN111444475B (en) | Fault-tolerant CKF filtering fusion method applied to flight test data analysis | |
| CN116338655A (en) | DVL speed error calibration method based on DMD-LSTM model | |
| CN116834977A (en) | Range control method for satellite orbit data | |
| CN116026325A (en) | Navigation method and related device based on neural process and Kalman filtering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |