CN102706347B - Inertial sensor network node device and information fusion method thereof - Google Patents
Inertial sensor network node device and information fusion method thereof Download PDFInfo
- Publication number
- CN102706347B CN102706347B CN201210151841.3A CN201210151841A CN102706347B CN 102706347 B CN102706347 B CN 102706347B CN 201210151841 A CN201210151841 A CN 201210151841A CN 102706347 B CN102706347 B CN 102706347B
- Authority
- CN
- China
- Prior art keywords
- unit
- optimal
- sensor
- depression
- information fusion
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses an inertial sensor network node device and an information fusion method thereof and belongs to the technical field of inertial navigation location. The network node device comprises an angle mount redundancy inertial measurement unit formed by an inertial sensor array (11) and an optimal information fusion system (22) based on a digital signal processing microprocessor. The optimal information fusion system (22) based on the digital signal processing microprocessor comprises a data collection unit (23), a reduced order processing unit (24) based on a virtual sensor, a system model, a measurement model building unit (25), a normalized form building unit (26) based on linear transformation, an optimal filter unit (27) and an inertial sensing network node optimal output unit (28). The information fusion method not only can obtain optimal error state estimation, but also can obtain the optimal output state estimation. The optimal output state can be directly used for inertial navigation resolving.
Description
Technical field
The present invention relates to a kind of inertial sensor network node device and information fusion method thereof, belong to inertial navigation field of locating technology.
Background technology
The navigation system of distributed inertial sensor network is adopted to be a kind of new Navigation System Design theory, it is the navigation sensor in low cost of new generation in recent years, small size, lightweight, as MEMS (MEMS) inertial sensor, MSIS (the solid-state inertial sensor of microminiature), optical fibre gyro, fibre optic accelerometer etc., and the new technology that the basis of the embedded microprocessor of high-speed high capacity and distributed modular electronic equipment grows up.The navigation system of distributed inertial sensor network can provide inertial data and other metrical information for motion carrier (as aircraft, naval vessel, spacecraft etc.), by navigation algorithm and information fusion algorithm, realize navigator fix, attitude determined, motion control, the various functions such as inertia aligning.
Based on the navigation system of distributed sensor networks, it is the distributed different spatial being configured in carrier of multiple sensor network nodes systems by having identical or different performance, function, each sensor network nodes is used for measuring the Local Navigation information of carrier, be made up of inertial sensor assembly and microprocessor module, all the sensors network node forms distributed navigation network structure completely jointly.Inertial sensor network node system is configured in multiple positions of carrier, the distributed measurement information of redundancy can not only be provided for the navigation of carrier, and be that the electronic equipment of carrier is as radar tracking, equipment load etc., the measurement system of local is provided, the local motion compensated inertial states information of carriers electron equipment can also be provided for simultaneously.Based on the navigation system of sensor network, by reconstructing and sharing limited computational resource, not only failure tolerant level can be improved, and the sensors configured network system function of dynamic.
Due to the inertial sensor of low cost, small size, lightweight and the precision of combination thereof and reliability on the low side, how while giving full play to inertial sensor advantage of new generation, improve system accuracy and reliability, and make system have good fault-tolerant ability, become the hot issue of research both at home and abroad.Current research shows, adopts sensor redundancy technology can improve certainty of measurement and reliability.By multiple independent inertial sensor composition array, form redundant sensor system, and by multi-sensor information fusion technology, effectively can realize the object of the performance being better than single-sensor.JPL laboratory as the current U.S. proposes " virtual gyroscope technology ", measures same angular speed improve certainty of measurement by N number of independently MEMS gyro.The researchers such as NORTHWEST CHINA polytechnical university, Zhejiang University have also carried out research to virtual gyro.
At present, for inertia redundant sensor system, both at home and abroad redundant configuration scheme, redundant sensor system reliability, fault diagnosis and fault-tolerant control etc. are carried out and can be studied, and propose some effective redundant configuration configurations.But current scheme normally to conduct a research for same type of sensor and designs, the sexual same error characteristics of general hypothesis sensor, the information fusion method adopted least square method substantially, or improving one's methods to least square method, as weighted least-squares method etc.But, there are different error characteristics etc. for dissimilar inertial sensor and identical type sensor, also there is no the effective optimum inertance network node apparatus based on multisensor redundant configuration, also there is no effective optimal information fusion processing method.
Summary of the invention
The present invention, in order to overcome the deficiency of existing inertia redundant apparatus on optimal performance, proposes a kind of inertial sensor network node device and information fusion method, for distributed inertance network navigation system.
The present invention adopts following technical scheme for solving its technical problem:
A kind of inertial sensor network node device, comprise the angle mount redundancy Inertial Measurement Unit that is made up of inertial sensor array and the optimal information fusion system based on Digital Signal Processing microprocessor, wherein inertial sensor array comprises m gyro, m is natural number, n accelerometer, n is natural number, optimal information fusion system based on Digital Signal Processing microprocessor comprises data acquisition unit, based on the depression of order processing unit of virtual-sensor, system model and measurement model construction unit, based on the standard type construction unit of linear transformation, optimal filter unit and the optimum output unit of inertia sensing network node, m gyro in inertial sensor array, n accelerometer, be connected respectively with based on the data acquisition unit in the optimal information fusion system of Digital Signal Processing microprocessor, based on the depression of order processing unit of virtual-sensor, system model and measurement model construction unit and be linked in sequence based on the standard type construction unit of linear transformation, depression of order processing unit based on virtual-sensor is connected with data acquisition unit respectively with optimal filter unit, optimal filter unit is connected with the optimum output unit of inertia sensing network node.
A kind of information fusion method of inertial sensor network node device, comprise the steps: all inertia measurement information that data acquisition unit obtains by (1), transfer in the depression of order processing unit based on virtual-sensor, first the error characteristics of each sensor are analyzed, then the metrical information composition subset of the similar inertial sensor of same error characteristic will be had, according to the installation matrix of inertial sensor in each subset, carry out depression of order process, build the Equivalent observation amount of depression of order;
(2) Equivalent observation amount step (1) obtained, in system model and measurement model construction unit, builds the error model of equal value of depression of order, and builds the measurement model of equal value of depression of order further; Based in the standard type construction unit of linear transformation, equivalence according to each subset of sensor of depression of order process installs matrix, adopt the linear transformation method based on kernel, by system model and measurement model, be converted into state equation and the observational equation of standard Kalman filtering;
(3) in optimal filter unit, adopt optimum Kalman filter to carry out Recursive Filtering and resolve, obtain optimum error state and estimate; In the optimum output unit of inertia sensing network node, according to the Optimal error state estimation of optimal filter unit, calculate optimum output state; This optimum output state is the optimal estimation of angular speed relative to installation shaft and acceleration, is directly used in inertial navigation and resolves.
Beneficial effect of the present invention is as follows:
1, angle mount redundant sensor allocation plan is adopted, devise a kind of new inertial sensor network node device, unlike the prior art yes, not only can adopt the inertial sensor of identical type, and dissimilar inertial sensor can be adopted, also can adopt the inertial sensor of same error characteristic and different error characteristics.
2, devise a kind of new optimal information integration method, not only can obtain optimum error state and estimate, also can obtain optimum output state and estimate.
3, devise a kind of depression of order treatment technology based on virtual-sensor, build the Equivalent observation amount of depression of order, and build depression of order error model of equal value and depression of order measurement model on this basis.By adopting depression of order treatment technology, the dimension of optimal filter can be reduced, significantly reducing the amount of calculation of information processing.
4, sensor-based installation matrix, adopts Zero Space Method and Its to carry out linear transformation, observation model can be converted to standard type.Based on the linear transformation method of kernel, not only should may be used for the system model of depression of order process, also can be directly used in the measurement model without depression of order process, build the Kalman filter of standard type.
5, this inertial sensor network node device, it is optimum exports as relative to the angular speed of installation shaft and acceleration, by the redundant measurement of the angle mount of multisensor, is treated to optimum normal orthogonal inertance element and exports, can be directly used in inertial navigation and resolve.Meanwhile, the estimation error of its optimum also can be used for compensation and the correction of sensor.
Accompanying drawing explanation
Fig. 1 is inertial sensor network node device structured flowchart of the present invention.
Fig. 2 is the depression of order process chart based on virtual-sensor of the present invention.
Fig. 3 is that depression of order error model of the present invention and measurement model build flow chart.
Fig. 4 is that the standard type based on linear transformation of the present invention builds flow chart.
Fig. 5 is that optimum Kalman filter of the present invention is resolved and optimum output flow chart.
Fig. 6 (a) is an inertial sensor network node device example of the present invention, and Fig. 6 (b) is the virtual bench of equal value of optimal information fusion process of the present invention.
Fig. 7 is an inertial sensor network node device simulated effect of the present invention.
Detailed description of the invention
Below in conjunction with accompanying drawing, the invention is described in further details.
As shown in Figure 1, inertial sensor network node device of the present invention comprises by inertial sensor array 11 and optimal information fusion system 22 two large divisions altogether based on DSP (Digital Signal Processing microprocessor).
Wherein, inertial sensor array is by m gyro, and m is natural number, n accelerometer, and n is natural number composition.Inertial sensor not only can adopt the inertial sensor of identical type, and can adopt dissimilar inertial sensor, also can adopt the inertial sensor of same error characteristic and different error characteristics.Such as inertial sensor can adopt MEMS (MEMS) gyro, mems accelerometer, optical fibre gyro, fibre optic accelerometer etc., and these dissimilar sensor combinations.Inertial sensor array, by the mode of angle mount redundancy, forms SRIMU (angle mount redundancy Inertial Measurement Unit).The metrical information of SRIMU, through the optimal information fusion system 22 based on DSP microprocessor, obtains optimum normal orthogonal inertance element and exports.
Based on the optimal information fusion system 22 of DSP by comprising data acquisition unit 23, depression of order processing unit 24, system model and the measurement model construction unit 25 based on virtual-sensor, the optimum output unit 28 of standard type construction unit 26, optimal filter unit 27 and the inertia sensing network node based on linear transformation.M gyro in inertial sensor array 11, m is natural number, n accelerometer, n is natural number, be connected respectively with based on the data acquisition unit 23 in the optimal information fusion system 22 of Digital Signal Processing microprocessor, based on virtual-sensor depression of order processing unit 24, system model and measurement model construction unit 25 and be linked in sequence based on the standard type construction unit 26 of linear transformation, depression of order processing unit 24 based on virtual-sensor is connected with data acquisition unit 23, and optimal filter unit 27 is connected with the optimum output unit 28 of inertia sensing network node.
This optimal information integration method adopts the microprocessor based on DSP to carry out the realization of blending algorithm, by the inertial sensor information obtained by data acquisition unit 23, by optimal information fusion algorithm, optimum fusion process is carried out to the metrical information of SRIMU, it is optimum exports as relative to the angular speed of installation shaft and acceleration information, can be directly used in inertial navigation and resolve.Meanwhile, the estimation error of its optimum also can be used for compensation and the correction of sensor.
Based on virtual-sensor depression of order handling process as shown in Figure 2, obtain all inertia measurement information through data acquisition unit 23, comprise the information of all m gyro and all n accelerometer, be designated as: z
1, z
2..., z
n(wherein, m+n=N).
The error characteristics of each sensor are analyzed, based on the thought of virtual-sensor, will metrical information composition L the subset of the similar inertial sensor of same error characteristic be had:
by the installation matrix M of each subset
l, build the Equivalent observation amount of depression of order
be respectively:
Wherein, M
i(i=1,2 ..., L) and be the installation matrix of i-th subset.By this depression of order process, the dimension of observed quantity can be reduced.Such as i-th subset comprises 10 gyros, all installs along same reference axis, then can adopt virtual sensor techniques, and these 10 gyros are equivalent to 1 virtual gyro, thus the observed quantity that 10 tie up is reduced to 1 dimension observed quantity of equal value; If these 10 gyros to be distributed in the different angles of X-Y plane, then these 10 gyros can be equivalent to 2 virtual gyros respectively along X-axis and Y-axis, thus the observed quantity that 10 tie up be reduced to 2 dimension observed quantities of equal value.
As shown in Figure 3, the depression of order process based on virtual-sensor obtains depression of order observed quantity of equal value
after, build the error model of equal value of depression of order:
Wherein,
for the systematic error state vector of k moment equivalence,
for the margin of error of k moment i-th subset; φ
k, k-1for the state-transition matrix of discrete system; W
eq, k-1for the system noise of equivalence; Γ
k-1for dynamic noise matrix.The measurement model of equal value of further structure depression of order is:
Ｚ
eq，k＝H
eq，kX
eq，k+M
eqU
k+V
eq，k(3)
Wherein,
for the state quantity measurement vector of k moment equivalence; H
eq, kfor the measurement matrix of equivalence; U
kfor the real motion state vector of carrier, the output state namely solved required for optimal information fusion system of the present invention, for the SRIMU of 6DOF, be three axis angular rates and 3-axis acceleration that inertance network node needs measurement; M
eqfor the installation matrix of virtual-sensor of equal value; V
eq, kfor measurement noise of equal value.
Due to equivalent system error model formula (2): X
eq, k=φ
k, k-1x
eq, k-1+ Γ
k-1w
eq, k-1modular form (3) is measured: Z with equivalence
eq, k=H
eq, kx
eq, k+ M
equ
k+ V
eq, kthe dynamical system formed, does not meet the standard type of optimum Kalman filter, and therefore the present invention adopts the linear transformation based on kernel technology, and this dynamical system is converted to standard type.Standard type based on linear transformation builds flow process as shown in Figure 4, according to the installation matrix M of equivalence
eq, build linear transformation battle array T based on Zero Space Method and Its
eq, wherein T
eqfor M
eqthe base of left kernel, namely meet:
T
eqM
eq＝0 (4)
And then, the measurement model of depression of order (3) can be turned to standard type:
T
eqZ
eq，k＝T
eqH
eq，kX
eq，k+T
eqV
eq，k(5)
So far, error model (2) and the measurement model (5) after changing, can form the kalman filter state equation and measurement equation that meet standard type, namely
X
k＝φ
k，k-1X
k-1+Γ
k-1W
k-1(6a)
Z
k＝H
kX
k+V
k(6b)
Optimum Kalman filter is resolved and optimum exports flow process as shown in Figure 5, according to the standard type kalman filter state equation (6a) after conversion and measurement equation (6b), adopt optimum Kalman's recurrence filter equation group, carry out the optimal estimation of error state.Its recurrence equation group is as follows:
By recurrence equation group, calculate the optimal estimation of error state
after, can calculate optimum output by following formula is further:
Wherein,
This optimum exports
be the optimal corner speed relative to installation shaft and acceleration, inertial navigation can be directly used in and resolve.Meanwhile, optimum estimation error
also can be used for compensation and the correction of sensor.
As shown in Fig. 6 (a), a kind of typical inertial sensor network node device example of design.This inertial sensor array adopts angle mount redundancy mounting means, and form SRIMU, its inertial sensor array is made up of 8 accelerometers and 2 gyros, and wherein 8 accelerometers are arranged in X-Y plane, and 2 gyro installations are at Z axis.
The error model of accelerometer 1 ~ 4 is identical, is random walk
and first-order Markov process
combination, namely
Wherein, τ
1for correlation time, w
b1and w
r1for white noise.
The error model of accelerometer 5 ~ 8 is identical, is random walk
namely
The error model of gyro 1 ~ 2 is identical, is first-order Markov process, namely
As shown in Fig. 6 (b), adopt optimal information integration method of the present invention, the similar inertial sensor composition gyro subset of same error model will be had, accelerometer subset 1, accelerometer subset 2, adopt based on after the depression of order process of virtual-sensor, obtain virtual gyro of equal value, virtual accelerometer 1, virtual accelerometer 2, thus the measurement amount that 10 tie up is reduced to 5 dimension measurement amounts of equal value, by 14 dimension error states, (accelerometer 1 ~ 4 respectively has 2 dimensions, accelerometer 5 ~ 8 respectively has 1 dimension, gyro 1 ~ 2 respectively has 1 dimension) reduce to 7 dimensions, therefore the dimension of optimal filter is reduced significantly, and then reduce amount of calculation.
When after reduced-order model, further according to the installation matrix of new virtual-sensor, adopt the linear transformation method of kernel, obtain kalman filter state equation and the observational equation of standard type, carry out filtering and resolve.By the optimal estimation of Kalman filter
the optimum output that further calculating is final.
Inertial sensor network node device simulated effect as shown in Figure 7, based on optimal information integration method of the present invention, devise the optimal information fusion system of the optimal information fusion system of gyro+accelerometer subset 1, the optimal information fusion system of gyro+accelerometer subset 2, all inertial sensor, and compare with adopting traditional least square method information fusion system.Suppose that simulation time is 300min, obtain X-axis acceleration output error (i.e. the difference of optimal information fusion output valve and nominal value) as shown in Figure 7.
As seen from Figure 7, inertial sensor network node device of the present invention and information fusion method thereof, not only can be applied to the sensor array of identical type sensor, same error model, and the sensor array (as accelerometer subset 1 and the combination of accelerometer subset 2) of dissimilar sensor array (as gyro combines from accelerometer) and different error model can be applied to.
Be it can also be seen that by Fig. 7, for the combination of all sensors array, adopt traditional least square information fusion, its Main Function is the weighted average of carrying out certain form, to the sensor of different error model (if accelerometer subset 1 is compared with accelerometer 2, there is good steady-state characteristic, but its random deviation is also larger at the beginning) optimum result can not be obtained; And optimal information integration method of the present invention, dissimilar error model can be carried out optimum fusion (no matter its error characteristics how), optimum result can be obtained.
Claims (1)
1. the information fusion method of an inertial sensor network node device, the device that the method uses comprises the angle mount redundancy Inertial Measurement Unit that is made up of inertial sensor array (11) and the optimal information fusion system (22) based on Digital Signal Processing microprocessor, wherein inertial sensor array (11) comprises m gyro, m is natural number, n accelerometer, n is natural number, optimal information fusion system (22) based on Digital Signal Processing microprocessor comprises data acquisition unit (23), based on the depression of order processing unit (24) of virtual-sensor, system model and measurement model construction unit (25), based on the standard type construction unit (26) of linear transformation, optimal filter unit (27) and the optimum output unit (28) of inertia sensing network node, m gyro in inertial sensor array (11), n accelerometer, be connected respectively with based on the data acquisition unit (23) in the optimal information fusion system (22) of Digital Signal Processing microprocessor, based on the depression of order processing unit (24) of virtual-sensor, system model and measurement model construction unit (25) and be linked in sequence based on the standard type construction unit (26) of linear transformation, depression of order processing unit (24) based on virtual-sensor is connected with data acquisition unit (23), optimal filter unit (27) is connected with the optimum output unit (28) of inertia sensing network node, it is characterized in that, this information fusion method, comprises the steps:
(1) all inertia measurement information data acquisition unit (23) obtained, transfer in the depression of order processing unit (24) based on virtual-sensor, first the error characteristics of each sensor are analyzed, then the metrical information composition subset of the similar inertial sensor of same error characteristic will be had, according to the installation matrix of inertial sensor in each subset, carry out depression of order process, build the Equivalent observation amount of depression of order;
(2) Equivalent observation amount step (1) obtained, in system model and measurement model construction unit (25), builds the error model of equal value of depression of order, and builds the measurement model of equal value of depression of order further; Based in the standard type construction unit (26) of linear transformation, equivalence according to each subset of sensor of depression of order process installs matrix, adopt the linear transformation method based on kernel, by system model and measurement model, be converted into state equation and the observational equation of standard Kalman filtering;
(3) in optimal filter unit (27), adopt optimum Kalman filter to carry out Recursive Filtering and resolve, obtain optimum error state and estimate; In the optimum output unit (28) of inertia sensing network node, according to the Optimal error state estimation of optimal filter unit (27), calculate optimum output state; This optimum output state is the optimal estimation of angular speed relative to installation shaft and acceleration, is directly used in inertial navigation and resolves.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210151841.3A CN102706347B (en) | 2012-05-17 | 2012-05-17 | Inertial sensor network node device and information fusion method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210151841.3A CN102706347B (en) | 2012-05-17 | 2012-05-17 | Inertial sensor network node device and information fusion method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102706347A CN102706347A (en) | 2012-10-03 |
| CN102706347B true CN102706347B (en) | 2015-03-25 |
Family
ID=46899299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210151841.3A Expired - Fee Related CN102706347B (en) | 2012-05-17 | 2012-05-17 | Inertial sensor network node device and information fusion method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102706347B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103148853B (en) * | 2013-03-20 | 2015-08-12 | 清华大学 | Based on satellite attitude determination method and the system of star sensor |
| CN103453904B (en) * | 2013-09-26 | 2015-12-02 | 哈尔滨工程大学 | A kind of redundancy configuration structure of Inertial Measurement Unit |
| US20170299388A9 (en) * | 2015-05-22 | 2017-10-19 | InvenSense, Incorporated | Systems and methods for synthetic sensor signal generation |
| US9797726B2 (en) * | 2015-08-10 | 2017-10-24 | GM Global Technology Operations LLC | Reduced-order fail-safe IMU system for active safety application |
| CN105424040B (en) * | 2016-01-15 | 2019-09-13 | 极翼机器人(上海)有限公司 | A kind of novel MEMS inertial sensor array redundant configurations method |
| CN107270894B (en) * | 2017-06-02 | 2020-11-06 | 南京理工大学 | GNSS/SINS deep integrated navigation system based on dimension reduction |
| CN107289934A (en) * | 2017-08-03 | 2017-10-24 | 重庆市皓颖科技发展有限公司 | A kind of attitude transducer of low-cost and high-precision |
| CN107421537B (en) * | 2017-09-14 | 2020-07-17 | 桂林电子科技大学 | Object motion attitude sensing method and system based on rigid body grid of inertial sensor |
| JP2019074322A (en) * | 2017-10-12 | 2019-05-16 | ソニー株式会社 | Information processing device, information processing method, and program |
| CN108225313B (en) * | 2017-12-29 | 2020-12-01 | 中国电子科技集团公司第十三研究所 | Attitude and heading reference instrument based on redundant MEMS sensor |
| CN109099910A (en) * | 2018-06-29 | 2018-12-28 | 广东星舆科技有限公司 | High Accuracy Inertial Navigation System and implementation method based on inertial navigation unit array |
| CN111121820B (en) * | 2019-12-16 | 2022-04-01 | 南京理工大学 | MEMS inertial sensor array fusion method based on Kalman filtering |
| CN113587920B (en) * | 2020-04-30 | 2024-02-20 | 阿里巴巴集团控股有限公司 | Motion measurement method, motion measurement device, electronic equipment and computer readable storage medium |
| CN112815937B (en) * | 2020-12-31 | 2023-08-01 | 中国人民解放军战略支援部队航天工程大学 | Optimal weight estimation method for data fusion of redundant inertial measurement unit |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB1536648A (en) * | 1976-10-27 | 1978-12-20 | Matra | Attitude control devices for space vehicles |
| CN102023010A (en) * | 2010-10-26 | 2011-04-20 | 西安理工大学 | MEMS (micro-electromechanical system)-based wavelet field multisensor information fusion system and fusion method |
-
2012
- 2012-05-17 CN CN201210151841.3A patent/CN102706347B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB1536648A (en) * | 1976-10-27 | 1978-12-20 | Matra | Attitude control devices for space vehicles |
| CN102023010A (en) * | 2010-10-26 | 2011-04-20 | 西安理工大学 | MEMS (micro-electromechanical system)-based wavelet field multisensor information fusion system and fusion method |
Non-Patent Citations (2)
| Title |
|---|
| Distributed data fusion algorithms for inertial network systems;D.J.Allerton et al.;《IET Radar Sonar Navig.》;20081231;第2卷(第1期);51-62 * |
| 虚拟陀螺技术在MEMS惯性导航系统中的应用;何昆鹏等;《哈尔滨工程大学学报》;20091031;第30卷(第10期);1123-1128 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102706347A (en) | 2012-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102706347B (en) | Inertial sensor network node device and information fusion method thereof | |
| CN107525503B (en) | Adaptive cascade Kalman filtering method based on combination of dual-antenna GPS and MIMU | |
| CN106017470B (en) | Micro inertial measurement unit screening technique and combined type micro-inertia measuring device | |
| CN103323007B (en) | A kind of robust federated filter method based on time-variable measurement noise | |
| CN109029448B (en) | Monocular vision inertial positioning's IMU aided tracking model | |
| CN105371868A (en) | Error calibration and compensation method for accelerometer unit of inertially stabilized platform system | |
| CN103712598B (en) | Attitude determination method of small unmanned aerial vehicle | |
| CN103148853B (en) | Based on satellite attitude determination method and the system of star sensor | |
| CN106979781A (en) | High-precision Transfer Alignment based on distributed inertance network | |
| CN109931955A (en) | Strapdown inertial navigation system Initial Alignment Method based on the filtering of state correlation Lie group | |
| CN102679977A (en) | Distributive navigation unit based on inertia network and information fusion method thereof | |
| CN105737834B (en) | A kind of Relative Navigation robust filtering method based on track mean element | |
| CN106767846B (en) | Three axis stabilized satellite does not have to the attitude acquisition method and system of gyro | |
| CN113175926B (en) | Self-adaptive horizontal attitude measurement method based on motion state monitoring | |
| CN104677355A (en) | Multi-sensor fusion based virtual gyroscope and method | |
| CN111993427B (en) | Self-stability-increasing control method, device, terminal, system and readable storage medium for aerial work robot | |
| CN102042833B (en) | Motion tracing method and system | |
| CN109916399B (en) | Carrier attitude estimation method under shadow | |
| CN110737194B (en) | Multi-MEMS sensor combination attitude measurement method | |
| CN113008229A (en) | Distributed autonomous integrated navigation method based on low-cost vehicle-mounted sensor | |
| Fu et al. | A novel optimal redundant inertial sensor configuration in strapdown inertial navigation system | |
| CN115659108A (en) | Acceleration calculation method | |
| CN112212860B (en) | Distributed filtering micro-nano satellite attitude determination method with fault tolerance | |
| CN114526729A (en) | Course optimization method of MEMS inertial positioning system based on redundancy technology | |
| Adnane et al. | Reliable Kalman filtering for satellite attitude estimation under gyroscope partial failure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150325 Termination date: 20170517 |