CN108759871A - A kind of strapdown inertial navigation system coarse alignment method based on improvement EMD Preprocessing Algorithms - Google Patents
A kind of strapdown inertial navigation system coarse alignment method based on improvement EMD Preprocessing Algorithms Download PDFInfo
- Publication number
- CN108759871A CN108759871A CN201810715581.5A CN201810715581A CN108759871A CN 108759871 A CN108759871 A CN 108759871A CN 201810715581 A CN201810715581 A CN 201810715581A CN 108759871 A CN108759871 A CN 108759871A
- Authority
- CN
- China
- Prior art keywords
- imf
- continuation
- signal
- coarse alignment
- emd
- 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.)
- Granted
Links
Classifications
-
- 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)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Navigation (AREA)
Abstract
The present invention proposes a kind of based on improvement empirical mode decomposition (Empirical Mode Decomposition, EMD) pretreated strapdown inertial navigation system (Strapdown Inertial Navigation System, SINS) coarse alignment method, this method is first with extreme learning machine (Extreme Learning Machine, ELM continuation) is carried out to the output signal of inertial sensor in SINS, the extreme point of required continuation obtains required whole continuation sequences, inhibits the influence of end effect；On this basis, intrinsic mode function (Intrinsic Mode Function, IMF) component of signal after continuation is calculated using EMD；It is finally based on the variable quantity that Shannon comentropies calculate the Shannon entropys of all adjacent IMF components, to complete the reconstruct of signal, to significantly improve the coarse alignment precision and robustness of system under the premise of not changing device precision.
Description
Technical field
That the present invention designs is a kind of strapdown inertial navigation system (Strapdown Inertial Navigation
System, SINS) coarse alignment method, more precisely, be utilize a kind of improved empirical mode decomposition method
(Empirical Mode Decomposition, EMD) pre-processes the output data of accuracy inertial sensor, thus
The precision and robustness of SINS coarse alignments are improved under the premise of not changing sensor accuracy.
Background technology
Now with the development of sensing technology, all kinds of carriers require its navigation system on the basis for meeting precision index
On, realize low cost, miniaturization and light-weight design, and there is stronger adaptive capacity to environment.Since its is small, weight
Gently, good concealment, capacity of will are strong, are not easy the advantages such as disturbed, and SINS is occupied an important position in each field.
Initial Alignment Technique is the premise of SINS normal works, and the speed and precision of alignment directly affect inertial navigation system
Quickly startup and navigation accuracy.Initial alignment is divided into coarse alignment and precision is accurate, and the purpose of wherein coarse alignment is thick in a short time
The initial strap-down matrix for calculating SINS slightly, to the basis smoothly carried out for fine alignment and guarantee.And analytic expression method is
Most common coarse alignment method, i.e., directly using accelerometer and gyro to the measurement of gravitational vectors and earth's spin vector come structure
Non-collinear vectors is built, the calculating of strapdown attitude matrix is realized by analytic expression method, to realize coarse alignment.Due to that can not use
The precision of the metrical information of any external information, accelerometer and gyro directly determines the precision of coarse alignment.However in reality
In use, the precision of Inertial Measurement Unit (Inertial Measurement Unit, IMU) is restricted by equipment volume.Cause
This, how to improve the performance of system under the premise of not changing device precision is one of the hot spot studied at present.
On the one hand, include a large amount of Random Drift Error in low precision IMU output signals in the limitation due to making industry,
Especially in a dynamic environment, Random Drift Error characteristic is more difficult to grasp, and can not carry out accurately estimating and compensating to it, from
And influence measurement accuracy of the IMU to carrier linear velocity and angular speed；On the other hand, fitful wind existing in actual use, hair
Motivation vibration, personnel walk about, the factors such as sea beat, SINS in the aligning process inevitably by various interference, to
Influence the precision and alignment speed of alignment.Therefore, it is to improve SINS coarse alignments to carry out noise suppression preprocessing to the output signal of IMU
The effective means of energy.
MEMS-IMU signal denoisings processing mode includes mainly Kalman filter method, wavelet analysis method, Empirical Mode at present
State decomposition method etc..The premise of Kalman filter optimal estimation is that system model is accurately known, and low precision IMU noises in practice
With stronger uncertainty, accurate modeling can not be carried out to system, to influence denoising effect；Denoising based on wavelet analysis
Method relies on the characteristics of its multiresolution to achieve certain effect in signal denoising field, but the quality of denoising effect is completely dependent on
In the selection of wavelet basis function, therefore this method has centainly restricted in practical applications；EMD methods are to rely on data sheet
The time scale of body carries out multiple adaptive decomposition to signal, and basic function need not be preset in decomposable process, therefore should
Method has greater advantage in terms of handling Dynamic Signal.However the end effect in EMD methods is to influence algorithm reliability
Principal element, presently, there are the end effect restrainable algorithms based on end effect, based on support vector machines end effect suppression
Algorithm processed, the end effect restrainable algorithms etc. based on Artificial neural network ensemble have the overall permanence, whole that can not show consideration for actual signal
The shortcomings of time-consuming for a process, real-time is poor.
For this purpose, the present invention propose it is a kind of based on improving the pretreated SINS coarse alignment methods of EMD, this method first with
Extreme learning machine (Extreme Learning Machine, ELM) carries out continuation, the pole of required continuation to the output signal of IMU
Value point obtains required whole continuation sequences, inhibits the influence of end effect；On this basis, believe after calculating continuation using EMD
Number intrinsic mode function (Intrinsic Mode Function, IMF) component；It is finally based on Shannon comentropies and calculates institute
The variable quantity for having the Shannon entropys of adjacent IMF components, to complete the reconstruct of signal.
Invention content
The object of the present invention is to provide it is a kind of based on improve EMD algorithms strapdown inertial navigation system coarse alignment method,
Under the premise of not changing device precision, the precision and robustness of SINS coarse alignments are improved.
The purpose of the present invention is what is realized by following steps：
Step 1：Fiber optic gyro strapdown inertial navigation system is installed on carrier, system is preheated and is acquired is each
The data of sensor；
Step 2：The influence that end effect is eliminated in continuation is carried out to collected IMU output signals using ELM；
Step 3：Data after continuation are decomposed into the form of a series of IMF components and remainder using EMD algorithms；
Step 4：IMU signals are reconstructed in the variable quantity for calculating the Shannon entropys of adjacent IMF components；
Step 5：Using analytic expression coarse alignment algorithm, the thick right of SINS is completed using the IMU output signals after pretreatment
Quasi- process, to improve the coarse alignment precision and robustness of SINS.
In the step 2 of the above method, continuation is carried out to collected IMU output signals using ELM and eliminates end effect
Influence, specific method is：
1) first by time series continuation to the right, using 7 adjacent data as the input of ELM, the right adjacent thereto (or
The left side) output of the data as ELM, in this, as training sample；
2) each new study carries out new prediction again after all one predicted value before this is added thereto, repeatedly
Constantly training study, required whole continuation sequences are obtained according to the extreme point of required continuation；
In the step 3 of the above method, using EMD algorithms by the data after continuation be decomposed into a series of IMF components with it is remaining
The form of item, specific method are：
1) assume that signal is x (t), it is m to take the sequence of lower envelope local mean value composition thereon1(t), then：
h1(t)=x (t)-m1(t)
For non-linear, Non-stationary Data, general single treatment is not enough to form IMF, some asymmetrical waves are still deposited
?.H1(t) regard pending data as and repeat aforesaid operations k times, obtain：
hk(t)=hk-1(t)-mk(t)
Work as hk(t) when meeting the condition of IMF, first IMF is just obtained, f is denoted as1(t)=hk(t)。
2) first IMF is separated from signal, obtains residual signal r1(t) it is：
r1(t)=x (t)-f1(t)
3) r1(t) it is used as signal to be decomposed, repetitive is above-mentioned 1)~and 3) step calculated, decomposes obtain successively：
Until residual signal rn(t) information in institute's research contents meaning very little or becomes a monotonic function, cannot
Until filtering out IMF again.So far, signal x (t) has been broken down into n IMF fi(t) (i=1,2 ..., n) and a remainder rn
(t) form of sum：
In the step 4 of the above method, the variable quantity for calculating the Shannon entropys of adjacent IMF components carries out weight to IMU signals
Structure, specific method are：
1) according to the definition of Shannon comentropies, the Shannon entropy for calculating each IMF components is Si(i=1,
2 ..., n)；
2) it is S according to the Shannon entropy of each IMFiThe converted quantity Δ of adjacent entropy is calculated in (i=1,2 ..., n)
Si=Si+1-Si(i=1,2 ..., n-1), and find maximum variable quantity；
3) the concrete numerical value j of the IMF components corresponding to Shannon entropy maximum variable quantities is determineds：
4) in order to reduce being mixed into for noise, j after selective stackingsA IMF components and remainder, the reconstruct of complete pair signals：
In the step 5 of the above method, using analytic expression coarse alignment algorithm, specific method is：
1) non-colinear under local gravity vector sum rotational-angular velocity of the earth 2 inertial coodinate systems (i) of information architecture is utilized
Vector：
WhereinIt is local geographic latitude, ωieFor rotational-angular velocity of the earth；
2) it and then is utilizingWithConstitute a new vectorI.e.：According to inertial coodinate system to carrier
Transition matrix between coordinate system (b)Following relationship can be obtained：
Therefore, it can solve
It 3) can be by the attitude matrix of b systems to navigational coordinate system (n) according to chain ruleIt is divided into 3 parts：
WhereinCoordinate conversion matrix for terrestrial coordinate system (e) with respect to n systems,For i systems square is converted with respect to the coordinate of e systems
Battle array：
Δ t is the sampling interval.Therefore initial strap-down matrix can be obtained, analytic expression coarse alignment process is completed.
Advantage of the invention is that：(1) pre- using the progress denoising of the IMU output signals of the low precision of signal processing centering
Processing completes the coarse alignment of SINS, the essence of system can be greatly improved under the premise of not improving device precision on this basis
Degree and robustness；(2) signal series is prolonged using extreme learning machine method when being pre-processed to IMU sensor signals
It opens up, eliminates end effect problem when EMD noise reductions, improve the precision of noise suppression preprocessing algorithm；(3) simultaneously, improved method utilizes
Shannon information entropy theories, the signal after being decomposed to EMD are reconstructed, and further suppress sensor noise and external environment
Interference, improve the robustness of preprocess method.
Description of the drawings
Fig. 1 is the method for the present invention flow diagram；
Fig. 2 is improved EMD Preprocessing Algorithms flow diagram；
Fig. 3 is original gyroscope and accelerometer signal power spectral density plot；
Fig. 4 is to utilize the gyroscope and accelerometer signal power spectral density plot improved after EMD algorithms pre-process.
Specific implementation mode
Below in conjunction with specific implementation case, the present invention is described in detail.
The present invention provides a kind of based on the strapdown inertial navigation system coarse alignment method for improving EMD Preprocessing Algorithms, side
Method schematic diagram is as depicted in figs. 1 and 2.The purpose of the present invention is what is realized by following steps：
1, fiber optic gyro strapdown inertial navigation system on carrier is installed, system is fully warmed-up, is started to work,
And acquire the data of each sensor；
2, first by time series continuation to the right, using 7 adjacent data as the input of ELM, the right adjacent thereto (or
The left side) output of the data as ELM, in this, as training sample；New study is all the pre- of one before this every time
Measured value carries out new prediction again after being added thereto, and constantly training study, obtains according to the extreme point of required continuation repeatedly
Required whole continuation sequences；End effect is eliminated to carry out continuation to collected IMU output signals using ELM methods
It influences；
3, to being handled using the IMU signals (setting signal as x (t)) after ELM method continuation, lower envelope part thereon is taken
The sequence of mean value composition is m1(t), then：
h1(t)=x (t)-m1(t)
For non-linear, Non-stationary Data, general single treatment is not enough to form IMF, some asymmetrical waves are still deposited
?.H1(t) regard pending data as and repeat aforesaid operations k times, obtain：
hk(t)=hk-1(t)-mk(t)
Work as hk(t) when meeting the condition of IMF, first IMF is just obtained, f is denoted as1(t)=hk(t)。
First IMF is separated from signal, obtains residual signal r1(t) it is：
r1(t)=x (t)-f1(t)
R1(t) it is used as signal to be decomposed, repetitive above-mentioned steps to be calculated, decomposes obtain successively：
Until residual signal rn(t) information in institute's research contents meaning very little or becomes a monotonic function, cannot
Until filtering out IMF again.So far, signal x (t) has been broken down into n IMF fi(t) (i=1,2 ..., n) and a remainder rn
(t) form of sum：
In this way, IMU output datas after continuation to be decomposed into the form of a series of IMF components and remainder using EMD algorithms；
4, according to the definition of Shannon comentropies, the Shannon entropy for calculating each IMF components is Si(i=1,
2 ..., n)；It is S according to the Shannon entropy of each IMFiThe converted quantity Δ S of adjacent entropy is calculated in (i=1,2 ..., n)i
=Si+1-Si(i=1,2 ..., n-1), and find maximum variable quantity；Determine the IMF corresponding to Shannon entropy maximum variable quantities
The concrete numerical value j of components：
In order to reduce being mixed into for noise, j after selective stackingsA IMF components and remainder, the reconstruct of complete pair signals：
In this way, the variable quantity based on Shannon entropys can complete the reconstruct to IMU signals；
5, using analytic expression coarse alignment algorithm, the coarse alignment mistake of SINS is completed using the IMU output signals after pretreatment
Journey.
It is sweared using the non-colinear under local gravity vector sum rotational-angular velocity of the earth 2 inertial coodinate systems (i) of information architecture
Amount：
WhereinIt is local geographic latitude, ωieFor rotational-angular velocity of the earth；
Then it is utilizingWithConstitute a new vectorI.e.：It is sat according to inertial coodinate system to carrier
Transition matrix between mark system (b)Following relationship can be obtained：
Therefore, it can solve
It can be by the attitude matrix of b systems to navigational coordinate system (n) according to chain ruleIt is divided into 3 parts：
WhereinCoordinate conversion matrix for terrestrial coordinate system (e) with respect to n systems,For i systems square is converted with respect to the coordinate of e systems
Battle array：
Wherein Δ t is the sampling interval.It is hereby achieved that initial strap-down matrix, completes analytic expression coarse alignment process.
The effect of the present invention can be verified by following track test：
Track test environment is built first, using the low-precision optical fiber gyro IMU of laboratory development, is fixed and is installed on
On the vibration isolation mounting device of instruction carriage boot, while high-precision optical fiber gyro IMU being installed as attitude reference, profit beside it
The output information of IMU, sample frequency 100Hz, total length of data 50min are acquired with reinforced notebook computer.
The power spectral density of collected IMU initial data is analyzed first, and is located in advance using the improvement EMD that the present invention is carried
Reason method pre-processes initial data, obtains the power spectral density after denoising, as shown in Figure 3 and Figure 4.It is basic herein
On, the coarse alignment of optical fiber IMU is completed using analytic expression coarse alignment algorithm, the alignment time is 2min.Then carried out using initial data
Analytic expression coarse alignment alignment result is：Pitching angle error is 0.0407 °, rolling angle error is -0.0136 °, course angle error be -
1.083°；It is using the coarse alignment result obtained based on the coarse alignment method for improving EMD preprocess methods designed by the present invention：
0.0279°,0.0053°,-0.7208°.Therefore, method provided by the invention has more accurate estimated accuracy and robustness,
The coarse alignment performance of SINS can be effectively improved.
Claims (4)
1. one kind is used based on the strapdown for improving empirical mode decomposition (Empirical Mode Decomposition, EMD) algorithm
Property navigation system (Strapdown Inertial Navigation System, SINS) coarse alignment method, mainly include it is following
Step：
Step 1：Fiber optic gyro strapdown inertial navigation system is installed on carrier, system is fully warmed-up, is started to work,
And acquire the data of each sensor；
Step 2：Limit of utilization learning machine (Extreme Learning Machine, ELM) is to collected optical fibre gyro inertia
Measuring unit (Inertial Measurement Unit, IMU) output signal carries out the influence that end effect is eliminated in continuation；
Step 3：The data after continuation are decomposed into a series of intrinsic mode functions (Intrinsic Mode using EMD algorithms
Function, IMF) component and remainder form；
Step 4：IMU signals are reconstructed in the variable quantity for calculating the Shannon entropys of adjacent IMF components；
Step 5：Using analytic expression coarse alignment algorithm, the coarse alignment mistake of SINS is completed using the IMU output signals after pretreatment
Journey, to improve the coarse alignment precision and robustness of SINS.
2. carrying out continuation elimination endpoint effect to collected IMU output signals using ELM according to claim 1 step 2
The influence answered, specific method are：
1) first by time series continuation to the right, using 7 adjacent data as the input of ELM, the right adjacent thereto (or it is left
Side) output of the data as ELM, in this, as training sample；
2) each new study carries out new prediction again after all one predicted value before this is added thereto, repeatedly constantly
Ground training study, required whole continuation sequences are obtained according to the extreme point of required continuation.
3. according to claim 1 step 3 using EMD algorithms by the data after continuation be decomposed into a series of IMF components with
The form of remainder, specific method are：
1) assume that signal is x (t), it is m to take the sequence of lower envelope local mean value composition thereon1(t), then：
h1(t)=x (t)-m1(t)
For non-linear, Non-stationary Data, general single treatment is not enough to form IMF, some asymmetrical waves still have.?
h1(t) regard pending data as and repeat aforesaid operations k times, obtain：
hk(t)=hk-1(t)-mk(t)
Work as hk(t) when meeting the condition of IMF, first IMF is just obtained, f is denoted as1(t)=hk(t)。
2) first IMF is separated from signal, obtains residual signal r1(t) it is：
r1(t)=x (t)-f1(t)
3) r1(t) it is used as signal to be decomposed, repetitive is above-mentioned 1)~and 3) step calculated, decomposes obtain successively：
Until residual signal rn(t) information in institute's research contents meaning very little or becomes a monotonic function, cannot screen again
Until going out IMF.So far, signal x (t) has been broken down into n IMF fi(t) (i=1,2 ..., n) and a remainder rn(t) sum
Form：
4. the variable quantity of the Shannon entropys of the adjacent IMF components of calculating according to claim 1 step 4 carries out IMU signals
Reconstruct, specific method are：
1) according to the definition of Shannon comentropies, the Shannon entropy for calculating each IMF components is Si(i=1,2 ..., n)；
2) it is S according to the Shannon entropy of each IMFi(i=1,2 ..., n) the converted quantity Δ S of adjacent entropy is calculatedi=
Si+1-Si(i=1,2 ..., n-1), and find maximum variable quantity；
3) the concrete numerical value j of the IMF components corresponding to Shannon entropy maximum variable quantities is determineds：
4) in order to reduce being mixed into for noise, j after selective stackingsA IMF components and remainder, the reconstruct of complete pair signals：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810715581.5A CN108759871B (en) | 2018-07-03 | 2018-07-03 | Improved EMD preprocessing algorithm-based strapdown inertial navigation system coarse alignment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810715581.5A CN108759871B (en) | 2018-07-03 | 2018-07-03 | Improved EMD preprocessing algorithm-based strapdown inertial navigation system coarse alignment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108759871A true CN108759871A (en) | 2018-11-06 |
| CN108759871B CN108759871B (en) | 2021-01-05 |
Family
ID=63975749
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810715581.5A Active CN108759871B (en) | 2018-07-03 | 2018-07-03 | Improved EMD preprocessing algorithm-based strapdown inertial navigation system coarse alignment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108759871B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109443393A (en) * | 2018-12-11 | 2019-03-08 | 中国人民解放军火箭军工程大学 | A kind of inertial navigation method for extracting signal and system based on blind separation algorithm |
| CN109855653A (en) * | 2019-03-08 | 2019-06-07 | 哈尔滨工程大学 | A kind of scaling method after the noise reduction process of redundance type MEMS-IMU |
| CN110220711A (en) * | 2019-05-22 | 2019-09-10 | 北京化工大学 | A kind of piston-mode motor shock characteristic extracting method based on EMD |
| CN111680581A (en) * | 2020-05-22 | 2020-09-18 | 电子科技大学 | Improved LCD mechanical fault feature extraction method for extreme learning endpoint continuation |
| CN112858485A (en) * | 2021-01-15 | 2021-05-28 | 沈阳工业大学 | Acoustic emission diagnosis method for rotor rub-impact fault |
| CN112902950A (en) * | 2021-01-21 | 2021-06-04 | 武汉大学 | Novel initial alignment method for MEMS-level IMU in low-speed motion carrier |
| CN114088077A (en) * | 2021-12-10 | 2022-02-25 | 哈尔滨工业大学 | Improved hemispherical resonator gyro signal denoising method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629243A (en) * | 2012-03-02 | 2012-08-08 | 燕山大学 | End effect suppression method based on neural network ensemble and B-spline empirical mode decomposition (BS-EMD) |
| CN103440226A (en) * | 2013-06-26 | 2013-12-11 | 燕山大学 | EMD (Empirical Mode Decomposition) endpoint effect suppression method based on HMM (Hidden Markov Model) correction and neural network extension |
| CN104573248A (en) * | 2015-01-16 | 2015-04-29 | 东南大学 | EMD based fiber-optic gyroscope temperature drift multi-scale extreme learning machine training method |
| CN106529680A (en) * | 2016-10-27 | 2017-03-22 | 天津工业大学 | Multiscale extreme learning machine integrated modeling method based on empirical mode decomposition |
| CN107063306A (en) * | 2017-04-14 | 2017-08-18 | 东南大学 | A kind of optical fibre gyro vibration compensation algorithm based on improved EEMD and arrangement entropy |
| CN107607835A (en) * | 2017-09-12 | 2018-01-19 | 国家电网公司 | A kind of transmission line of electricity laser ranging Signal denoising algorithm based on improvement EEMD |
-
2018
- 2018-07-03 CN CN201810715581.5A patent/CN108759871B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629243A (en) * | 2012-03-02 | 2012-08-08 | 燕山大学 | End effect suppression method based on neural network ensemble and B-spline empirical mode decomposition (BS-EMD) |
| CN103440226A (en) * | 2013-06-26 | 2013-12-11 | 燕山大学 | EMD (Empirical Mode Decomposition) endpoint effect suppression method based on HMM (Hidden Markov Model) correction and neural network extension |
| CN104573248A (en) * | 2015-01-16 | 2015-04-29 | 东南大学 | EMD based fiber-optic gyroscope temperature drift multi-scale extreme learning machine training method |
| CN106529680A (en) * | 2016-10-27 | 2017-03-22 | 天津工业大学 | Multiscale extreme learning machine integrated modeling method based on empirical mode decomposition |
| CN107063306A (en) * | 2017-04-14 | 2017-08-18 | 东南大学 | A kind of optical fibre gyro vibration compensation algorithm based on improved EEMD and arrangement entropy |
| CN107607835A (en) * | 2017-09-12 | 2018-01-19 | 国家电网公司 | A kind of transmission line of electricity laser ranging Signal denoising algorithm based on improvement EEMD |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109443393A (en) * | 2018-12-11 | 2019-03-08 | 中国人民解放军火箭军工程大学 | A kind of inertial navigation method for extracting signal and system based on blind separation algorithm |
| CN109855653A (en) * | 2019-03-08 | 2019-06-07 | 哈尔滨工程大学 | A kind of scaling method after the noise reduction process of redundance type MEMS-IMU |
| CN110220711A (en) * | 2019-05-22 | 2019-09-10 | 北京化工大学 | A kind of piston-mode motor shock characteristic extracting method based on EMD |
| CN111680581A (en) * | 2020-05-22 | 2020-09-18 | 电子科技大学 | Improved LCD mechanical fault feature extraction method for extreme learning endpoint continuation |
| CN112858485A (en) * | 2021-01-15 | 2021-05-28 | 沈阳工业大学 | Acoustic emission diagnosis method for rotor rub-impact fault |
| CN112858485B (en) * | 2021-01-15 | 2024-04-19 | 沈阳工业大学 | Acoustic emission diagnosis method for rotor rub-impact fault |
| CN112902950A (en) * | 2021-01-21 | 2021-06-04 | 武汉大学 | Novel initial alignment method for MEMS-level IMU in low-speed motion carrier |
| CN112902950B (en) * | 2021-01-21 | 2022-10-21 | 武汉大学 | Initial alignment method for MEMS-level IMU in low-speed motion carrier |
| CN114088077A (en) * | 2021-12-10 | 2022-02-25 | 哈尔滨工业大学 | Improved hemispherical resonator gyro signal denoising method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108759871B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108759871A (en) | A kind of strapdown inertial navigation system coarse alignment method based on improvement EMD Preprocessing Algorithms | |
| CN109974697A (en) | A kind of high-precision mapping method based on inertia system | |
| CN101706284B (en) | Method for increasing position precision of optical fiber gyro strap-down inertial navigation system used by ship | |
| CN105698822B (en) | Initial Alignment Method between autonomous type inertial navigation based on reversed Attitude Tracking is advanced | |
| CN110057354B (en) | Geomagnetic matching navigation method based on declination correction | |
| US11519731B2 (en) | Pedestrian adaptive zero-velocity update point selection method based on a neural network | |
| CN100547352C (en) | The ground speed testing methods that is suitable for fiber optic gyro strapdown inertial navigation system | |
| CN105509739A (en) | Tightly coupled INS/UWB integrated navigation system and method adopting fixed-interval CRTS smoothing | |
| CN101183004A (en) | Method for online real-time removing oscillation error of optical fibre gyroscope SINS system | |
| CN103869379A (en) | Magnetometer correcting method with optimized and modified BP neural network based on genetic algorithm | |
| CN103076026B (en) | A kind of method determining Doppler log range rate error in SINS | |
| CN103776449B (en) | A kind of initial alignment on moving base method that improves robustness | |
| CN108458709B (en) | Airborne distributed POS data fusion method and device based on vision-aided measurement | |
| CN103557856A (en) | Random drift real-time filtering method for fiber-optic gyroscope | |
| CN110779532A (en) | Geomagnetic navigation system and method applied to near-earth orbit satellite | |
| CN112197765A (en) | Method for realizing fine navigation of underwater robot | |
| CN110887472B (en) | Polarization-geomagnetic information deep fusion fully-autonomous attitude calculation method | |
| CN113503892A (en) | Inertial navigation system moving base initial alignment method based on odometer and backtracking navigation | |
| CN104501809B (en) | Attitude coupling-based strapdown inertial navigation/star sensor integrated navigation method | |
| CN106153037A (en) | The indoor orientation method of a kind of robot, Apparatus and system | |
| CN108225375A (en) | A kind of optimization coarse alignment method of the anti-outer speed outlier based on medium filtering | |
| CN104634348B (en) | Attitude angle computational methods in integrated navigation | |
| CN107036595A (en) | Deformation of hull angular estimation method based on interacting multiple model filters | |
| CN114111840B (en) | DVL error parameter online calibration method based on integrated navigation | |
| CN115166856A (en) | Unmanned ship weight magnetic measurement method, system, equipment and computer readable storage medium |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Gao Wei Inventor after: Zhang Ya Inventor after: Yu Fei Inventor after: Wang Yanyan Inventor after: Wang Kai Inventor before: Zhang Ya Inventor before: Yu Fei Inventor before: Gao Wei Inventor before: Wang Yanyan Inventor before: Wang Kai |