CN103761267B - Kalman filtering system in comprehensive modular avionics system - Google Patents
Kalman filtering system in comprehensive modular avionics system Download PDFInfo
- Publication number
- CN103761267B CN103761267B CN201410004214.6A CN201410004214A CN103761267B CN 103761267 B CN103761267 B CN 103761267B CN 201410004214 A CN201410004214 A CN 201410004214A CN 103761267 B CN103761267 B CN 103761267B
- Authority
- CN
- China
- Prior art keywords
- module
- management module
- result
- results
- variance
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a Kalman filtering system in an integrated modular avionics system, which comprises a model management module, a Kalman filtering calculation module, an output result management module and an inquiry decomposition module, wherein the Kalman filtering calculation module performs Kalman filtering calculation according to model related data provided by the model management module, the output result management module stores iterative calculation results of the Kalman filtering calculation module, and the inquiry decomposition module sends inquiry requests to the result management module according to the requirements of subsystems in the avionics system so as to obtain required data. The invention uses the unified Kalman filtering system to provide Kalman filtering service for each subsystem, thereby reducing the code amount of the system, improving the reliability of the system to a certain extent and improving the utilization rate of the computing resources of the system.
Description
Technical field
The present invention relates to the integrated approach in a kind of avionics system, specifically, refer to a kind of comprehensive
Kalman filtering system and its implementation in modular avionics system.
Background technology
Comprehensively modularized aviation electronics (integrated modular avionics, ima) is the association type avionics framework that continues
New a kind of avionics framework afterwards.The feature of this framework is to achieve shared, the multiplexing of resource, thereby, significantly drops
The low cost of avionics system, weight.The realization of ima framework needs to carry out substantial amounts of synthesis, particularly function synthesized.Function is comprehensive
The Major Difficulties closing are that how to maximize realizes the multiplexing of functional module and the shared of information data.Avionics system
In, such as navigation subsystem, radar tracking subsystem, infrared track subsystem etc., all can use Kalman filtering function, every height
Typically all can there is each independent Kalman filtering module, these each independent modules aim at respective system service in system,
Can not substitute mutually from each other.Additionally, each Kalman filtering module can produce some intermediate results in running, this
A little results can be used for carrying out, such as motor-driven detection, filtering performance assessment etc., and these intermediate results are often dropped.
Content of the invention
The deficiency existing for above-mentioned prior art, the goal of the invention of the present invention is to provide a kind of comprehensively modularized aviation
Kalman filtering system in electronic system, solves subsystems during ima function synthesized and shares Kalman filtering
Module and in filtering data sharing problem.
The goal of the invention of the present invention is achieved through the following technical solutions:
A kind of Kalman filtering system in comprehensively modularized avionics system, comprises model management module, Kalman
Filtering computing module, output result management module, query decomposition module:
Described model management module is used for being managed collectively all models of Kalman filtering, and Kalman filtering computing module leads to
Cross and submit required model request to model management module, model management module will return corresponding model relevant data；
Kalman filtering computing module carries out Kalman filtering according to the model relevant data that model management module exports and changes
In generation, calculates；
Output result management module preserve Kalman filtering computing module iterative calculation result, provide query function and
Obtain required data for Kalman filtering computing module and query decomposition module；
Inquiry request is sent to result according to the demand of subsystem each in avionics system by described query decomposition module
Management module is to obtain required data.
According to features described above, described Kalman filtering computing module comprises state prediction module, measurement prediction module, measurement
Residual error module, state update module, status predication variance module, the variance module of measurement residual error, filtering gain module and state
Variance update module；
Described state prediction module is used for the prediction of state；
Described measurement prediction module is used for the prediction of measurement；
Described measurement residual error module is used for measuring residual error；
Described state update module is used for current estimated state；
Described status predication variance module is used for the variance of status predication；
Described measurement residual variance module is used for measuring the variance of residual error；
Described filtering gain module is used for filtering gain；
Described state variance update module is used for the variance of estimated state.
According to features described above, described output result management module includes and each submodule in Kalman filtering computing module
The one-to-one output result of block manages submodule, specifically comprises status predication results management module, measures the management that predicts the outcome
Module, measurement residual result management module, state update results management module, status predication variance results management module, measurement
Residual variance results management module, filtering gain results management mould, state variance renewal results management module are defeated, output result pipe
Reason module also includes filter result management module, and described filter result management module is used for managing Kalman filtering an iteration
Result in cycle, the main state including inputting in current iteration, measurement, state variance, model and current iteration output
State, variance.
According to features described above, the structure of the result data of described output result management module storage is: block of information number, every
Byte-sized, timestamp, used model, |input paramete, result data etc. shared by individual block of information, described information block comprises the time
Stamp, obtain model that this result used, obtain the byte-sized shared by |input paramete, each block of information that this result used
Deng.
According to features described above, query demand is decomposed into the subquery that certain order is constituted, presses by described query decomposition module
Each submodule that subquery is sent in corresponding output result management module order, to obtain required data, will looked into
Ask result while be sent to demander this result also can be sent to filter result management module and enter row cache.
According to features described above, the structure of the query demand that described query decomposition module receives includes: timestamp, guarantees the quality
Phase, used model, data class；The query demand structure that query decomposition module sends includes: timestamp, shelf-life, is made
Use model；After query decomposition module receives inquiry request, some row accordingly is formulated according to the data class in request
Inquiry.
According to features described above, the concrete query script of described query decomposition module comprises the following steps:
If in avionics system, certain subsystem sends inquiry request to query decomposition module, if the type of this inquiry request
It is once filtering request, then this inquiry request is decomposed into once filtering subquery sequence by query decomposition module, comprises following step
Rapid:
The first step, sends query demand to filter result management module, if results needed exists, jumps to the 11st step,
If not existing, go to second step；
Second step, filtering subquery sequence sends request to status predication results management module, if there is not results needed,
Then status predication management module is asked to state prediction module transmission processe, if there is results needed, result returns to inquiry
Decomposing module, and in order to be filtered the next step subquery of subquery sequence；
3rd step, query decomposition module to measurement predict the outcome management module send inquiry request, if there is not required knot
Really, then measurement predicts the outcome management module to measurement prediction module transmission processe request, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
4th step, query decomposition module sends inquiry request to measurement residual result management module, if there is not required knot
Really, then measurement residual result management module is asked to measurement residual error module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
5th step, query decomposition module sends inquiry request to status predication variance results management module, if there is not institute
Need result, then status predication variance results management module is asked to status predication variance module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
6th step, query decomposition module sends inquiry request to measurement residual variance results management module, if there is not institute
Need result, then measurement residual variance results management module is asked to measurement residual variance module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
7th step, query decomposition module sends inquiry request to filtering gain results management module, if there is not required knot
Really, then filtering gain results management module is asked to filtering gain module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
8th step, query decomposition module updates results management module to state variance and sends inquiry request, if there is not institute
Need result, then state variance updates results management module and asks to state variance update module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
9th step, query decomposition module updates results management module to state and sends inquiry request, if there is not required knot
Really, then state updates results management module to the request of state update module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
Tenth step, state filtering result is sent to filter result management module and enters row cache；
11st step, returning result；
If in avionics system, certain subsystem sends inquiry request to query decomposition module, if the type of this inquiry request
It is the request in addition to once filtering request, then query decomposition module directly converts the request into corresponding subquery, accordingly
Subquery sends inquiry request to corresponding data management module, if corresponding data management module has results needed, ties
Fruit returns to query decomposition module, if there is not results needed, query decomposition module returns inquiry status of fail to task.
According to features described above, in each step, judge whether that results needed specifically includes following steps:
A) output result management module is searched in output result according to the used model in query demand and |input paramete
Whether there is the result meeting query demand;
If b) having the result meeting demand, the timestamp in timestamp and query demand in comparative result, see result
Whether exceed the shelf-life;
If c) there being multiple qualified results, the up-to-date result of selection result timestamp is exported.
Compared with prior art, make the beneficial effects of the present invention is changing subsystems in comprehensively modularized avionics
With the state of oneself special Kalman filtering module, provide karr using unified Kalman filtering system for subsystems
Graceful filtering service, reduces the size of code of system, improves the reliability of system to a certain extent, improves system-computed money
The utilization rate in source.
Brief description
Fig. 1 is the structural representation of the present invention；
Fig. 2 is the flow chart in filtering iteration cycle of query decomposition module of the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings the present invention is made a detailed description.
As shown in figure 1, the Kalman filtering system in a kind of comprehensively modularized avionics system of the present invention, comprise model
Management module, Kalman filtering computing module, output result management module, query decomposition module.
Described model management module is used for being managed collectively all models of Kalman filtering, and Kalman filtering computing module leads to
Cross and submit required model request to model management module, model management module will return corresponding model relevant data.Mould
Type management module, is used to manage some Filtering Model that may use in Kalman filtering.This module inputs as types of models
Label and model initialization parameter, are output as corresponding Filtering Model.The model that generally Kalman filtering may use has constant speed
Model, constant accelerator model, singer model and "current" model etc..Because Kalman filtering is linear filtering, so these moulds
Type can be described in the way of matrix.Store the corresponding matrix template of these models, then the parameter according to input
Required Filtering Model just can be generated.
Kalman filtering computing module carries out Kalman filtering according to the model relevant data that model management module exports and changes
In generation, calculates.Kalman filtering function is by sending corresponding data according to specified order to disparate modules and processing request
Mode finally to realize.Kalman filtering computing module comprises state prediction module, measurement prediction module, measurement residual error mould
Block, state update module, status predication variance module, the variance module of measurement residual error, filtering gain module and state variance are more
New module.
These modules are the requisite functional modules of Kalman filtering functional realiey.
Hypothesis current time is k, and corresponding pattern number is m, and the corresponding model matrix of Filtering Model m is respectively state transfer square
Battle array f, input matrix g, process noise covariance matrix q, measurement noise variance matrix r, these model matrixs can be by model
Management module input model number and corresponding model parameter and obtain.Input, output and the work(of each functional module are described below
Can:
A the input of () state prediction module is current estimated state and pattern number m, estimated state is designated as x(k |
K), a(i here | j) represent the value of a in i moment estimating in the j moment.The function of state prediction module is to input shape according to current
State x (k) and model, predict subsequent time, the i.e. state in k+1 moment, are designated as x(k+1 | k) wherein
X (k+1 | k)=f (k) x(k | k)+g (k) u (k).
(b) measure prediction module input be corresponding pattern number m and in the k moment state to the k+1 moment prediction, that is,
x(k+1|k).The function of state prediction module is the measured value in model m and x (k+1 | k) the prediction k+1 moment according to input, i.e. z
(k+1 | k) wherein
z(k+1|k)=h(k+1)x(k+1|k).
C input that () measures residual error module is the measured value z (k+1) and z (k+1 | k) of subsequent time.Measurement residual error module
Output be the k+1 moment measurement residual error v (k+1) wherein
v(k+1)=z(k+1)-z(k+1|k).
D the input of () status predication variance module is model and current time state variance p (k), output is in current k
Carve the prediction to k+1 moment state variance, i.e. and p (k+1 | k) wherein
p(k+1|k)=f(k)p(k)f(k)’+q(k).
(e) measure residual variance module input be model and in the current k moment prediction p to k+1 moment state variance
(k+1|k).Output is the variance that the k+1 moment measures residual error, is designated as s (k+1) wherein
s(k+1)=r(k+1)+h(k+1)p(k+1|k)h(k+1)’.
F the input of () filtering gain module is model and measures residual variance s (k+1) and when the current k moment is to k+1
Carve the prediction p (k+1 | k) of state variance.Output be k+1 moment filtering gain w (k+1) wherein
w(k+1)=p(k+1|k)h(k+1)’s(k+1)-1.
G the input of () state update module is to increase in the prediction x (k+1 | k) of the state to the k+1 moment in the k moment, k+1 moment
Benefit and the measurement residual error in k+1 moment.Output is the estimation of the state in the k+1 moment to the k+1 moment, is designated as x (k+1 | k+1) wherein
x(k+1|k+1)=x(k+1|k)+w(k+1)v(k+1).
H the input of () state variance update module is the residual variance s (k+ in the gain w (k+1) in k+1 moment, k+1 moment
1) and the k+1 moment gain w (k+1).Output is variance p (k+1) of the state estimated by the k+1 moment
p(k+1)=p(k+1|k)-w(k+1)s(k+1)w(k+1)’.
Described output result management module is used for managing the result that karr overflows filtering computing module output, comprises: state
Predict the outcome management module；Measure the management module that predicts the outcome；Measurement residual result management module；State updates results management mould
Block；Status predication variance results management module；Measurement residual variance results management module；Filtering gain results management mould；State
It is defeated that variance updates results management module.The corresponding output result of each submodule in Kalman filtering computing module manages mould
The submodule of block.Each output result manages submodule and has a data buffer queue for storing the knot of respective modules output
Really.Data buffer storage queue can be the queue of round-robin queue or linked list type.Data buffer storage queue only cached in the range of certain time
Data, time range can carry out to each output result management module specifying or dynamic in running as the case may be
State is changed, and exceeding the data in the time range specified can be removed automatically from buffer queue.
Each output result manages submodule and is responsible for preserving the output result of respective modules in certain period of time, output knot
Fruit management module can provide query function to obtain required data for other modules from accordingly result mould management module, including knot
The lookup of fruit, deletion and parameter is sent to corresponding function module and starts corresponding functional module and enter row operation and then return meter
Calculate result etc..
The output result of each submodule of Kalman filtering computing module is being sent to corresponding output result management mould
After the submodule of block, this data block can be stored in respective team after data is increased with some extraneous information blocks by results management module
In row, these increasedd block of informations include: timestamp (for the output time of record data), obtain what this result was used
Model, obtain byte-sized shared by |input paramete, each block of information that this result used etc..Store in results management module
The structure of result data be: block of information number, byte-sized, timestamp shared by each block of information, used model, input ginseng
Number, result data etc..When query demand is sent to results management module, the parameter in query demand mainly using model,
|input paramete, timestamp, shelf-life etc..The query demand structure being sent to results management module includes: timestamp, the shelf-life,
Model, |input paramete.Whether results management module searches in queue according to the used model in query demand and |input paramete
Have the result meeting query demand, if there being the result meeting demand, in the timestamp and query demand in comparative result when
Between stab, see whether result exceedes the shelf-life, if there being multiple qualified results, the up-to-date result of selection result timestamp is entered
Row output.Though if not finding the result meeting demand or find result but result have passed through the shelf-life, results management mould
Model in query demand and |input paramete are sent to corresponding functional module by block, and startup function module is calculated, and treats work(
Energy module calculates after finishing, and result is sent the sender of query demand.
Described query decomposition module is used to the module that the query demand being sent to Kalman filtering system is decomposed,
This module by query demand be decomposed into one graded constitute subquery, and order subquery is sent to corresponding result pipe
Reason module is to obtain required data.The structure of the query demand that data interaction and Task-decomposing module receive includes: the time
Stamp, shelf-life, used model, data class etc..The query demand structure that query decomposition module sends includes: timestamp, guarantor
Matter phase, used model.After query decomposition module receives inquiry request, if being formulated according to the data class in request
Arrange corresponding subquery.
Also comprise filter result management module in described output result management module, once change for managing Kalman filtering
For the result in the cycle, the main state including inputting in current iteration, measurement, state variance, model and current iteration are defeated
The state going out and variance.This module is similar with other management modules to be suitable in the storage organizations such as queue caching certain time
Filter result.
If the query demand that query decomposition module receives is if the output knot in an iteration cycle of Kalman filtering
Really, then the corresponding subquery of this query demand can make a look up first in filter result management module.If not finding requisite number
According to a series of subquery then can be carried out to obtain results needed, this result while Query Result is sent to demander
Also filter result management module can be sent to and enter row cache.When filter result management module receives the data needing caching, meeting
Accordingly compared with avoid data repeat store.
As shown in Figure 2, in avionics system, certain subsystem is to query decomposition for the course of work in filtering iteration cycle
Module sends inquiry request, if the type of this inquiry request is once filtering request, query decomposition module is by this inquiry request
It is decomposed into once filtering subquery sequence.
The first step, sends query demand to filter result management module, if results needed exists, jumps to the 11st step,
If not existing, go to second step.
Second step, filtering subquery sequence sends request to status predication results management module, if there is not results needed,
Then status predication management module is asked to state prediction module transmission processe, if there is results needed, result returns to inquiry
Decomposing module, and in order to be filtered the next step subquery of subquery sequence.
3rd step, query decomposition module to measurement predict the outcome management module send inquiry request, if there is not required knot
Really, then measurement predicts the outcome management module to measurement prediction module transmission processe request, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
4th step, query decomposition module sends inquiry request to measurement residual result management module, if there is not required knot
Really, then measurement residual result management module is asked to measurement residual error module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
5th step, query decomposition module sends inquiry request to status predication variance results management module, if there is not institute
Need result, then status predication variance results management module is asked to status predication variance module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
6th step, query decomposition module sends inquiry request to measurement residual variance results management module, if there is not institute
Need result, then measurement residual variance results management module is asked to measurement residual variance module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
7th step, query decomposition module sends inquiry request to filtering gain results management module, if there is not required knot
Really, then filtering gain results management module is asked to filtering gain module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
8th step, query decomposition module updates results management module to state variance and sends inquiry request, if there is not institute
Need result, then state variance updates results management module and asks to state variance update module transmission processe, if there is required knot
Really, then result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
9th step, query decomposition module updates results management module to state and sends inquiry request, if there is not required knot
Really, then state updates results management module to the request of state update module transmission processe, if there is results needed, result returns
To query decomposition module, and in order to be filtered the next step subquery of subquery sequence.
Tenth step, state filtering result is sent to filter result management module and enters row cache.
11st step, returning result.
If certain task sends inquiry request to query decomposition module, if the type of this inquiry request is except once filtering request
Request in addition, such as status predication request, measurement predictions request, measurement residual error request etc., then query decomposition module directly will be asked
Ask and be converted to corresponding subquery, corresponding subquery sends inquiry request, corresponding data to corresponding data management module
If management module has results needed, result returns to query decomposition module, if there is not results needed, query decomposition mould
If block certain subsystem in avionics system returns inquiry status of fail.
Claims (4)
1. the Kalman filtering system in a kind of comprehensively modularized avionics system, comprises model management module, Kalman's filter
Ripple computing module, output result management module, query decomposition module it is characterised in that:
Described model management module is used for being managed collectively all models of Kalman filtering, Kalman filtering computing module pass through to
Model management module submits required model request to, and model management module will return corresponding model relevant data；
Kalman filtering computing module carries out Kalman filtering iteration meter according to the model relevant data that model management module exports
Calculate, comprise state prediction module, measurement prediction module, measurement residual error module, state update module, status predication variance module,
The variance module of measurement residual error, filtering gain module and state variance update module；
Described state prediction module is used for the prediction of state；
Described measurement prediction module is used for the prediction of measurement；
Described measurement residual error module is used for measuring residual error；
Described state update module is used for current estimated state；
Described status predication variance module is used for the variance of status predication；
Described measurement residual variance module is used for measuring the variance of residual error；
Described filtering gain module is used for filtering gain；
Described state variance update module is used for the variance of estimated state；
Output result management module preserves the iterative calculation result of Kalman filtering computing module, provides query function and for card
Kalman Filtering computing module and query decomposition module obtain required data, include with Kalman filtering computing module in each
The one-to-one output result of submodule manages submodule, specifically comprises status predication results management module, measurement predicts the outcome
Management module, measurement residual result management module, state update results management module, status predication variance results management module,
Measurement residual variance results management module, filtering gain results management mould, state variance renewal results management module are defeated, output knot
Fruit management module also includes filter result management module, and described filter result management module is used for managing Kalman filtering once
Result in iteration cycle, the main state including inputting in current iteration, measurement, state variance, model and current iteration
The state of output, variance；
Query demand is decomposed into certain order according to the demand of subsystem each in avionics system by described query decomposition module
Subquery is sent to each submodule in corresponding output result management module to obtain by subquery in order that constitute
The data needing, while Query Result is sent to demander, this Query Result also can be sent to filter result management module and enters
Row cache, concrete query script comprises the following steps:
If in avionics system, certain subsystem sends inquiry request to query decomposition module, if the type of this inquiry request is one
Secondary filtering is asked, then this inquiry request is decomposed into once filtering subquery sequence by query decomposition module, comprises the steps of
The first step, sends query demand to filter result management module, if results needed exists, jumps to the 11st step, if not
Exist, go to second step；
Second step, filtering subquery sequence sends request to status predication results management module, if there is not results needed, shape
State prediction management module is asked to state prediction module transmission processe, if there is results needed, result returns to query decomposition
Module, and in order to be filtered the next step subquery of subquery sequence；
3rd step, query decomposition module to measurement predict the outcome management module send inquiry request, if there is not results needed,
Measurement predicts the outcome management module to measurement prediction module transmission processe request, if there is results needed, result returns to be looked into
Ask decomposing module, and in order to be filtered the next step subquery of subquery sequence；
4th step, query decomposition module sends inquiry request to measurement residual result management module, if there is not results needed,
To measurement residual error module transmission processe request, if there is results needed, result returns to be looked into measurement residual result management module
Ask decomposing module, and in order to be filtered the next step subquery of subquery sequence；
5th step, query decomposition module sends inquiry request to status predication variance results management module, if there is not required knot
Really, then status predication variance results management module is asked to status predication variance module transmission processe, if there is results needed,
Result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
6th step, query decomposition module sends inquiry request to measurement residual variance results management module, if there is not required knot
Really, then measurement residual variance results management module is asked to measurement residual variance module transmission processe, if there is results needed,
Result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
7th step, query decomposition module sends inquiry request to filtering gain results management module, if there is not results needed,
Filtering gain results management module is asked to filtering gain module transmission processe, if there is results needed, result returns to be looked into
Ask decomposing module, and in order to be filtered the next step subquery of subquery sequence；
8th step, query decomposition module updates results management module to state variance and sends inquiry request, if there is not required knot
Really, then state variance updates results management module to the request of state variance update module transmission processe, if there is results needed,
Result returns to query decomposition module, and in order to be filtered the next step subquery of subquery sequence；
9th step, query decomposition module updates results management module to state and sends inquiry request, if there is not results needed,
State updates results management module asks to state update module transmission processe, if there is results needed, result returns to be looked into
Ask decomposing module, and in order to be filtered the next step subquery of subquery sequence；
Tenth step, state filtering result is sent to filter result management module and enters row cache；
11st step, returning result；
If in avionics system, certain subsystem sends inquiry request to query decomposition module, if the type of this inquiry request is to remove
Once filter the request beyond asking, then query decomposition module directly converts the request into corresponding subquery, and son is looked into accordingly
Ask and send inquiry request to corresponding data management module, if corresponding data management module has results needed, result is returned
Return to query decomposition module, if there is not results needed, query decomposition module returns inquiry status of fail to task.
2. the Kalman filtering system in a kind of comprehensively modularized avionics system according to claim 1, its feature
The structure being the result data of described output result management module storage is: block of information number, byte shared by each block of information
Size, timestamp, used model, |input paramete, result data, described information block comprises timestamp, obtains this result and made
Model, obtain the byte-sized shared by |input paramete, each block of information that this result used.
3. the Kalman filtering system in a kind of comprehensively modularized avionics system according to claim 1, described looks into
The structure asking the query demand that decomposing module receives includes: timestamp, shelf-life, used model, data class；Inquiry point
The query demand structure that solution module sends includes: timestamp, shelf-life, used model；Query decomposition module receives inquiry
After request, a series of corresponding subqueries are formulated according to the data class in request.
4. the Kalman filtering system in a kind of comprehensively modularized avionics system according to claim 1, its feature
It is in each step of query decomposition module to judge whether that results needed specifically includes following steps:
A) whether output result management module searches in output result according to the used model in query demand and |input paramete
There is the result meeting query demand；
If b) having the result meeting demand, the timestamp in timestamp and query demand in comparative result, whether see result
Exceed the shelf-life；
If c) there being multiple qualified results, the up-to-date result of selection result timestamp is exported.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410004214.6A CN103761267B (en) | 2014-01-06 | 2014-01-06 | Kalman filtering system in comprehensive modular avionics system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410004214.6A CN103761267B (en) | 2014-01-06 | 2014-01-06 | Kalman filtering system in comprehensive modular avionics system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103761267A CN103761267A (en) | 2014-04-30 |
| CN103761267B true CN103761267B (en) | 2017-01-25 |
Family
ID=50528504
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410004214.6A Active CN103761267B (en) | 2014-01-06 | 2014-01-06 | Kalman filtering system in comprehensive modular avionics system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103761267B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112241254B (en) * | 2019-07-16 | 2024-02-02 | 北京车和家信息技术有限公司 | Code statistics method and device for Internet of vehicles multi-terminal system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201365232Y (en) * | 2009-01-15 | 2009-12-16 | 北京航空航天大学 | Kalman filter based on field programmable gate array |
| US8019736B1 (en) * | 2007-07-25 | 2011-09-13 | Rockwell Collins, Inc. | Systems and methods for combining a plurality of terrain databases into one terrain database |
| CN103366058A (en) * | 2013-07-06 | 2013-10-23 | 北京航空航天大学 | FPGA-based high-dimensional Kalman filter and implementation method thereof |
-
2014
- 2014-01-06 CN CN201410004214.6A patent/CN103761267B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8019736B1 (en) * | 2007-07-25 | 2011-09-13 | Rockwell Collins, Inc. | Systems and methods for combining a plurality of terrain databases into one terrain database |
| CN201365232Y (en) * | 2009-01-15 | 2009-12-16 | 北京航空航天大学 | Kalman filter based on field programmable gate array |
| CN103366058A (en) * | 2013-07-06 | 2013-10-23 | 北京航空航天大学 | FPGA-based high-dimensional Kalman filter and implementation method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103761267A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104123369B (en) | A kind of implementation method of the configuration management Database Systems based on graphic data base | |
| CN105426467B (en) | A kind of SQL query method and system for Presto | |
| CN108171971A (en) | Vehicular real time monitoring method and system based on Spark Streaming | |
| KR100991316B1 (en) | Use of simulation to generate predictions pertaining to a manufacturing facility | |
| CN109074284A (en) | For increasing and decreasing the method and system and computer program product of resource in proportion | |
| US20040186829A1 (en) | Database system, server, query posing method, and data updating method | |
| CN111881358B (en) | Object recommendation system, method and device, electronic equipment and storage medium | |
| CN108986872B (en) | Multi-granularity attribute weight Spark method for big data electronic medical record reduction | |
| CN107783985A (en) | A kind of distributed networks database query method, apparatus and management system | |
| CN108009270A (en) | A kind of text searching method calculated based on distributed memory | |
| CN104090934A (en) | Standard service platform distributed parallel computing database and retrieval method thereof | |
| CN106228263A (en) | Materials stream informationization methods based on big data | |
| CN105741070B (en) | A kind of method of the quick difference update storage of the real-time freight space freight rate result data of magnanimity | |
| CN105138686B (en) | A kind of instant application process for multistage storage data | |
| CN110309171A (en) | Data base query method, server and system | |
| US20160125095A1 (en) | Lightweight temporal graph management engine | |
| CN103761267B (en) | Kalman filtering system in comprehensive modular avionics system | |
| CN107729500A (en) | A kind of data processing method of on-line analytical processing, device and background devices | |
| CN109885620A (en) | Metadata read method and device based on Hive data warehouse | |
| CN106789163B (en) | Method, device and system for monitoring power utilization information of network equipment | |
| US11003431B2 (en) | Generating predictive metrics for virtualized deployments | |
| CN105045906A (en) | Estimation method and device of click rate of delivery information | |
| US10380135B2 (en) | Data aggregation and reporting environment for data center infrastructure management | |
| CN106992901A (en) | Method and apparatus for scheduling of resource simulated pressure | |
| Smolij | Management conception designer preproduction of electronic vehicles |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220523 Address after: 200241 Building 5, No. 432, ziyue Road, Minhang District, Shanghai Patentee after: Shanghai avionics Co.,Ltd. Address before: 432 Guiping Road, Xuhui District, Shanghai, 2003 Patentee before: CHINA AERONAUTICAL RADIO ELECTRONICS Research Institute |
|
| TR01 | Transfer of patent right |