CN111238462B - LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering - Google Patents
LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering Download PDFInfo
- Publication number
- CN111238462B CN111238462B CN202010060569.2A CN202010060569A CN111238462B CN 111238462 B CN111238462 B CN 111238462B CN 202010060569 A CN202010060569 A CN 202010060569A CN 111238462 B CN111238462 B CN 111238462B
- Authority
- CN
- China
- Prior art keywords
- fiber
- clustering
- optic gyroscope
- distribution
- automatic encoder
- 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 32
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 238000013507 mapping Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 4
- 239000000835 fiber Substances 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 12
- 238000012804 iterative process Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 239000013307 optical fiber Substances 0.000 abstract description 2
- 230000008859 change Effects 0.000 description 8
- 230000006872 improvement Effects 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C19/00—Gyroscopes; Turn-sensitive devices using vibrating masses; Turn-sensitive devices without moving masses; Measuring angular rate using gyroscopic effects
- G01C19/58—Turn-sensitive devices without moving masses
- G01C19/64—Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams
- G01C19/72—Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams with counter-rotating light beams in a passive ring, e.g. fibre laser gyrometers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C19/00—Gyroscopes; Turn-sensitive devices using vibrating masses; Turn-sensitive devices without moving masses; Measuring angular rate using gyroscopic effects
- G01C19/58—Turn-sensitive devices without moving masses
- G01C19/64—Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams
- G01C19/72—Gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams with counter-rotating light beams in a passive ring, e.g. fibre laser gyrometers
- G01C19/721—Details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Optics & Photonics (AREA)
- Electromagnetism (AREA)
- Power Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Gyroscopes (AREA)
Abstract
The invention discloses an LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering, which comprises the following steps: acquiring temperature and zero-bias data of the fiber-optic gyroscope to construct a training data set, and training the denoising automatic encoder layer by layer; constructing a depth automatic encoder based on the trained denoising automatic encoder; mapping the input x to obtain an embedded point z based on a depth automatic encoder; calculating soft distribution of an embedded point z and a clustering center, and constructing auxiliary target distribution; iteration is carried out between the calculation of the auxiliary objective function and the minimization of the kl divergence by taking the kl divergence of the soft distribution and the auxiliary objective distribution as an objective function, and parameters of the depth automatic encoder and a clustering center are updated; and segmenting according to the clustering result, and training by using an LSTM neural network on each segment to obtain the fiber-optic gyroscope temperature compensation model. The invention can realize the temperature compensation of the zero offset error of the gyroscope output, obtains good fitting and predicting effects and higher temperature environment adaptability, and improves the precision of the optical fiber gyroscope product.
Description
Technical Field
The invention relates to the technical field of fiber optic gyroscopes, in particular to an LSTM fiber optic gyroscope temperature compensation modeling method based on deep embedded clustering.
Background
The influence of temperature is one of main factors for restricting the performance of the fiber-optic gyroscope, when the temperature of a working environment changes, zero drift of an output signal of the gyroscope can be caused, and the performance of the gyroscope needs to be improved by performing temperature compensation on the zero drift. At present, some machine learning algorithms are applied to the establishment of a gyro temperature compensation model, such as a support vector machine, a wavelet neural network, an RBF neural network and the like.
A temperature compensation method of a fiber optic gyroscope based on a wavelet neural network is disclosed in the temperature compensation system of the fiber optic gyroscope based on the wavelet neural network of piezoelectric and acousto-optic, a stretching function and a translation function are introduced to carry out zero-offset modeling on the fiber optic gyroscope, and the approaching capability, the convergence rate and the fault-tolerant capability of a nonlinear function are better realized; a fiber optic gyroscope temperature compensation method based on an improved support vector machine is disclosed in a fiber optic gyroscope temperature drift compensation method of the improved support vector machine in the infrared and laser engineering. A genetic algorithm optimized network parameter GA-BP neural network temperature compensation model is disclosed in the optical fiber gyroscope temperature compensation based on the GA-BP neural network of the instrument technology and sensor.
The problems existing in the prior art are as follows:
1. the fiber-optic gyroscope is influenced by the natural temperature rise of the inertia element and the circuit board after being electrified, and the change situation of the internal temperature is complex. The gyro zero offset is not only influenced by temperature to change, but also related to the characteristic change of some key devices on the signal acquisition board along with the temperature change, and belongs to the system-level error; the ideal compensation effect is difficult to realize by the existing machine learning model.
2. The machine learning method is suitable for the premise that samples are independently and identically distributed, namely the samples accord with the same physical model, but the performance of the gyroscope is changed under different temperature conditions, and the gyroscope cannot be described by a single model, namely the gyroscope needs to be subjected to piecewise fitting; however, the piecewise regression generally artificially divides the piecewise interval according to the local characteristics of the signal, and introduces artificial errors.
Disclosure of Invention
Aiming at the defect of artificial segmentation fitting of a fiber optic gyroscope temperature compensation model in the prior art, the invention provides an LSTM fiber optic gyroscope temperature compensation modeling method based on deep embedded clustering.
The invention discloses an LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering, which comprises the following steps:
acquiring temperature and zero-bias data of the fiber-optic gyroscope to construct a training data set, and training the denoising automatic encoder layer by layer;
constructing a depth automatic encoder based on the trained denoising automatic encoder;
mapping an input x to obtain an embedded point z based on the depth automatic encoder;
calculating soft distribution of an embedded point z and a clustering center, and constructing auxiliary target distribution;
iteration is carried out between the calculation of the auxiliary objective function and the minimization of the kl divergence by taking the kl divergence of the soft distribution and the auxiliary objective distribution as an objective function, and parameters of the depth automatic encoder and a clustering center are updated;
and segmenting according to the clustering result, and training by using an LSTM neural network on each segment to obtain the fiber-optic gyroscope temperature compensation model.
As a further improvement of the invention, the layer-by-layer training denoising automatic encoder comprises:
inputting x after dropout into a coding layer of the denoising automatic coder to obtain an intermediate quantity h, and inputting h after dropout into a decoding layer of the automatic coder to obtain an output y;
and learning the denoising automatic encoder layer by layer, and transmitting the intermediate quantity h into the next denoising automatic encoder.
As a further improvement of the present invention, the depth automatic encoder is configured to include:
and connecting the coding layers of all the trained denoising automatic encoders, and connecting the decoding layers of all the denoising automatic encoders by a symmetrical structure to obtain the depth automatic encoder.
As a further improvement of the present invention, the depth automatic encoder is based on mapping an input x to obtain an embedded point z; the method comprises the following steps:
taking a coding part of the depth automatic coder as an initial mapping;
and mapping the input x to a feature space to obtain an embedded point z based on the initial mapping.
As a further improvement of the invention, the soft allocation of the embedding point z and the clustering center is calculated, and auxiliary target allocation is constructed; the method comprises the following steps:
using student T distribution as a kernel function, measuring the similarity of the embedded point z and the clustering center point mu, and obtaining soft distribution of the embedded point z and the clustering center;
and taking the result with the highest confidence level in the soft distribution to construct auxiliary target distribution.
As a further improvement of the invention, in the iterative process:
and (5) continuously iterating and optimizing until the clustering result of the point less than tol% between two clusters is changed, and considering that the clustering is stable and the iteration is stopped.
As a further improvement of the invention, the training of the LSTM neural network on each segment to obtain the fiber optic gyroscope temperature compensation model comprises the following steps:
taking temperature, temperature rate, and the product of temperature and temperature rate as input;
taking the zero offset of the current fiber-optic gyroscope as output;
and training the LSTM neural network to obtain a segmented temperature compensation model.
Compared with the prior art, the invention has the beneficial effects that:
the invention adopts a data-driven unsupervised learning method to finish the data clustering of the fiber-optic gyroscope under different temperature conditions, and provides data support for the establishment of a segmented temperature compensation model; the LSTM network is used for establishing a temperature compensation model, historical information of temperature change is fully utilized, the influence of a complex temperature field on inertial navigation output can be more accurately described, and the temperature compensation precision is improved; compared with a solution method for improving hardware lifting precision, the method is low in cost and easy to implement, and good fitting and prediction effects and high temperature environment adaptability can be obtained.
Drawings
FIG. 1 is a flowchart of an LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
The invention is described in further detail below with reference to the attached drawing figures:
the invention provides an LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering, which comprises the following steps: deep embedding clustering is carried out according to the characteristics of gyro data, and unsupervised learning is adopted for carrying out data clustering (clustering of fiber optic gyro data under different temperature conditions); segmenting the fiber-optic gyroscope data according to the clustering result, designing an LSTM network on each segment to construct a temperature compensation model, constructing a training set and a testing set according to experimental data, completing model training, obtaining a mathematical model of zero offset and temperature of the fiber-optic gyroscope, and predicting and compensating the zero offset error. The method can fit the current zero offset of the fiber-optic gyroscope according to the current temperature data and the temperature data of a period of time in the past, realize the temperature compensation of the zero offset error of the gyroscope output, obtain good fitting and predicting effects and higher temperature environment adaptability, and improve the precision of the fiber-optic gyroscope product.
As shown in fig. 1, the method of the present invention specifically includes:
step 1, collecting temperature and zero-offset data of a fiber optic gyroscope to construct a training data set, and training a denoising automatic encoder layer by layer; wherein,
the training method of the denoising automatic encoder comprises the following steps:
the noise encoder is composed of an encoding layer and a decoding layer, wherein x is input into the encoding layer after dropout to obtain an intermediate quantity h, and h is input into the decoding layer after dropout to obtain an output y; and learning the denoising automatic encoder layer by layer, and transmitting the intermediate quantity h into the next denoising automatic encoder.
Step 2, constructing a depth automatic encoder based on the trained denoising automatic encoder, and finely adjusting to reduce reconstruction loss; wherein,
the construction method of the depth automatic encoder comprises the following steps:
and connecting the coding layers of all the trained denoising automatic encoders, and connecting the decoding layers of all the denoising automatic encoders by a symmetrical structure to obtain the depth automatic encoder.
And 3, taking a coding part of the depth automatic coder as initial mapping, and mapping the input x into a feature space to obtain an embedded point z based on the initial mapping.
Step 4, calculating soft distribution of an embedded point z and a clustering center, and constructing auxiliary target distribution; the method specifically comprises the following steps:
using student T distribution as a kernel function, measuring the similarity of the embedded point z and the clustering center point mu, and obtaining soft distribution of the embedded point z and the clustering center;
and taking the result with the highest confidence level in the soft distribution, and constructing the auxiliary target distribution.
Step 5, iteration is carried out between the calculation of the auxiliary objective function and the minimization of the kl divergence by taking the kl divergence of the soft distribution and the auxiliary objective distribution as an objective function, and parameters and a clustering center of the depth automatic encoder are updated; wherein,
the parameters to be optimized comprise parameters in a deep neural network (automatic encoder) and clustering centers in clustering, the iterative optimization is continued until clustering results of points less than tol% between two clustering changes, and the clustering is considered to be stable and stops iteration.
Step 6, segmenting according to clustering results, and training each segment by using an LSTM neural network to obtain a fiber optic gyroscope temperature compensation model; wherein,
and training the LSTM network by taking the temperature, the temperature change rate and the product of the temperature and the temperature change rate at 10 sampling points as input and taking the zero offset of the current fiber-optic gyroscope as output to obtain a segmented temperature compensation model.
The invention has the advantages that:
the invention adopts a data-driven unsupervised learning method to finish the data clustering of the fiber-optic gyroscope under different temperature conditions, and provides data support for the establishment of a segmented temperature compensation model; the LSTM network is used for establishing a temperature compensation model, historical information of temperature change is fully utilized, the influence of a complex temperature field on inertial navigation output can be more accurately described, and the temperature compensation precision is improved; compared with a solution method for improving hardware lifting precision, the method is low in cost and easy to implement, and good fitting and prediction effects and high temperature environment adaptability can be obtained.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedding clustering is characterized by comprising the following steps:
acquiring temperature and zero-bias data of the fiber-optic gyroscope to construct a training data set, and training the denoising automatic encoder layer by layer;
constructing a depth automatic encoder based on the trained denoising automatic encoder;
mapping an input x to obtain an embedded point z based on the depth automatic encoder;
calculating soft distribution of an embedded point z and a clustering center, and constructing auxiliary target distribution;
iteration is carried out between the calculation of the auxiliary objective function and the minimization of the kl divergence by taking the kl divergence of the soft distribution and the auxiliary objective distribution as an objective function, and parameters of the depth automatic encoder and a clustering center are updated;
and segmenting according to the clustering result, and training by using an LSTM neural network on each segment to obtain the fiber-optic gyroscope temperature compensation model.
2. The method of claim 1, wherein said layer-by-layer training denoising auto-encoder comprises:
inputting x after dropout into a coding layer of the denoising automatic coder to obtain an intermediate quantity h, and inputting h after dropout into a decoding layer of the automatic coder to obtain an output y;
and learning the denoising automatic encoder layer by layer, and transmitting the intermediate quantity h into the next denoising automatic encoder.
3. The method of claim 1, wherein constructing the depth autoencoder comprises:
and connecting the coding layers of all the trained denoising automatic encoders, and connecting the decoding layers of all the denoising automatic encoders by a symmetrical structure to obtain the depth automatic encoder.
4. The method of claim 1, wherein said depth auto-encoder based, mapping an input x to an embedding point z; the method comprises the following steps:
taking a coding part of the depth automatic coder as an initial mapping;
and mapping the input x to a feature space to obtain an embedded point z based on the initial mapping.
5. The method of claim 1, wherein the soft assignments of the embedding point z and the cluster center are computed, an auxiliary target assignment is constructed; the method comprises the following steps:
using student T distribution as a kernel function, measuring the similarity of the embedded point z and the clustering center point mu, and obtaining soft distribution of the embedded point z and the clustering center;
and taking the result with the highest confidence level in the soft distribution to construct auxiliary target distribution.
6. The method of claim 1, wherein in an iterative process:
and (5) continuously iterating and optimizing until the clustering result of the point less than tol% between two clusters is changed, and considering that the clustering is stable and the iteration is stopped.
7. The method of claim 1, wherein the training using the LSTM neural network on each segment to obtain the fiber optic gyroscope temperature compensation model comprises:
taking temperature, temperature rate, and the product of temperature and temperature rate as input;
taking the zero offset of the current fiber-optic gyroscope as output;
and training the LSTM neural network to obtain a segmented temperature compensation model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010060569.2A CN111238462B (en) | 2020-01-19 | 2020-01-19 | LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010060569.2A CN111238462B (en) | 2020-01-19 | 2020-01-19 | LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111238462A CN111238462A (en) | 2020-06-05 |
| CN111238462B true CN111238462B (en) | 2021-12-28 |
Family
ID=70871400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010060569.2A Active CN111238462B (en) | 2020-01-19 | 2020-01-19 | LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111238462B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111895986A (en) * | 2020-06-30 | 2020-11-06 | 西安建筑科技大学 | MEMS gyroscope original output signal noise reduction method based on LSTM neural network |
| CN112067015B (en) * | 2020-09-03 | 2022-11-22 | 青岛歌尔智能传感器有限公司 | Step counting method and device based on convolutional neural network and readable storage medium |
| CN114061559B (en) * | 2021-11-16 | 2023-05-16 | 湖北三江航天万峰科技发展有限公司 | Compensation method, system and computer storage medium for zero offset drift of fiber optic gyroscope |
| CN114459455B (en) * | 2021-12-24 | 2023-02-14 | 浙江大学 | LSTM-based fiber-optic gyroscope scale factor error compensation method |
| CN118133604B (en) * | 2024-02-01 | 2024-09-13 | 北京林业大学 | Numerical simulation method, system, equipment and medium applying telescopic function |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107621269A (en) * | 2016-07-15 | 2018-01-23 | 北京计算机技术及应用研究所 | Fiber Optic Gyroscope Temperature Drift error compensating method |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007037926A2 (en) * | 2005-09-23 | 2007-04-05 | Sharp Laboratories Of America, Inc. | Mems pixel sensor |
| CN1865852A (en) * | 2006-06-27 | 2006-11-22 | 北京航空航天大学 | Output error compensating process for open-loop optic fiber gyroscope based on neural network |
| CN100510633C (en) * | 2007-02-08 | 2009-07-08 | 北京航空航天大学 | Neural net based temperature compensating optical fibre gyroscope |
| CN101566483B (en) * | 2009-05-22 | 2012-03-14 | 哈尔滨工程大学 | Vibration error compensation method of fiber optic gyro strap-down inertia measurement system |
| US10215587B2 (en) * | 2012-05-18 | 2019-02-26 | Trx Systems, Inc. | Method for step detection and gait direction estimation |
| CN103499345B (en) * | 2013-10-15 | 2016-04-20 | 北京航空航天大学 | A kind of Fiber Optic Gyroscope Temperature Drift compensation method based on wavelet analysis and BP neural network |
| CN103593538B (en) * | 2013-11-28 | 2017-03-22 | 东南大学 | Fiber optic gyroscope temperature drift modeling method by optimizing dynamic recurrent neural network through genetic algorithm |
| CN103954300A (en) * | 2014-04-30 | 2014-07-30 | 东南大学 | Fiber optic gyroscope temperature drift error compensation method based on optimized least square-support vector machine (LS-SVM) |
| CN104330095A (en) * | 2014-10-24 | 2015-02-04 | 中国兵器工业集团第二一四研究所苏州研发中心 | MEMS gyro temperature compensation method on basis of sectional fitting |
| CN104820757B (en) * | 2015-05-18 | 2018-02-06 | 中国电子科技集团公司第二十研究所 | MEMS gyro temperature drift characteristic neural network modeling approach based on physical model |
| CN105333888B (en) * | 2015-11-26 | 2018-02-23 | 湖北三江航天红峰控制有限公司 | It is a kind of to utilize a temperature experiment while the method for compensated optical fiber gyro constant multiplier and zero bias |
| CN106597852A (en) * | 2016-12-27 | 2017-04-26 | 中国船舶重工集团公司第七0五研究所 | MEMS gyroscope temperature compensation method based on RBF neural network |
| CN107832193B (en) * | 2017-11-13 | 2021-05-25 | 湖北三江航天红峰控制有限公司 | Based on I2Multi-board-card state monitoring method for C bus |
| EP3539728A1 (en) * | 2018-03-17 | 2019-09-18 | Tata Consultancy Services Limited | System and method for fault detection in robotic actuation |
-
2020
- 2020-01-19 CN CN202010060569.2A patent/CN111238462B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107621269A (en) * | 2016-07-15 | 2018-01-23 | 北京计算机技术及应用研究所 | Fiber Optic Gyroscope Temperature Drift error compensating method |
Non-Patent Citations (1)
| Title |
|---|
| Sklearn参数详解—聚类算法;张俊红;《https://blog.csdn.net/junhongzhang/article/details/103686878》;20180707;第1-13页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111238462A (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111238462B (en) | LSTM fiber-optic gyroscope temperature compensation modeling method based on deep embedded clustering | |
| CN112257341A (en) | Customized product performance prediction method based on heterogeneous data difference compensation fusion | |
| CN113158470B (en) | Oil reservoir automatic history fitting system and method based on transfer learning | |
| CN112235043B (en) | Distributed optical fiber abnormal data restoration device based on self-adaptive long-term and short-term memory | |
| CN111967688A (en) | Power load prediction method based on Kalman filter and convolutional neural network | |
| CN114463932B (en) | Non-contact construction safety distance active dynamic identification early warning system and method | |
| CN111832227B (en) | Shale gas saturation determination method, device and equipment based on deep learning | |
| CN108764475A (en) | The Gyro Random error compensating method and system of genetic wavelet neural network | |
| Fei et al. | Research on tunnel engineering monitoring technology based on BPNN neural network and MARS machine learning regression algorithm | |
| CN108563895B (en) | Interval model correction method for gear assembly body considering correlation | |
| CN115374375A (en) | Track clustering method, system and storage medium based on deep learning | |
| CN115471679A (en) | Method and intelligent system for synchronously assimilating water level and flow of natural river | |
| CN112836789A (en) | Ground connection wall deformation dynamic prediction method based on composite neural network algorithm | |
| CN110260914B (en) | Engineering safety monitoring system region division method based on measuring point space-time characteristics | |
| CN117827326A (en) | Remote sensing mapping data precision enhancement method and system | |
| CN116993832B (en) | CNN-based position prediction method and device for flight target | |
| CN117874448A (en) | Space-time weather prediction method based on transducer | |
| CN116628836A (en) | Data-driven supercritical airfoil performance prediction method, system and medium | |
| CN114996947A (en) | Three-dimensional oil reservoir numerical simulation method | |
| CN114004315A (en) | Method and device for incremental learning based on small sample | |
| US20240054348A1 (en) | Systems and method for masked multi-step multivariate time series power forcasting and estimation | |
| CN111504305B (en) | Underwater acoustic communication-MEMS sensor collaborative navigation positioning method and system | |
| CN117633494B (en) | Coal mine earth surface deformation prediction method based on AWC-LSTM model | |
| CN118296095B (en) | Method and system for determining regional boundary in homeland space planning | |
| CN118410710B (en) | Method for reducing influence of overburden on mineral exploration potential field data processing inversion |
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 |