CN108592914A - The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario - Google Patents
The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario Download PDFInfo
- Publication number
- CN108592914A CN108592914A CN201810305986.1A CN201810305986A CN108592914A CN 108592914 A CN108592914 A CN 108592914A CN 201810305986 A CN201810305986 A CN 201810305986A CN 108592914 A CN108592914 A CN 108592914A
- Authority
- CN
- China
- Prior art keywords
- geomagnetic
- inspecting robot
- point
- matching
- moment
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/04—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means
- G01C21/08—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means involving use of the magnetic field of the earth
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
Abstract
This hair discloses the complex region inspecting robot location navigation and time service method under a kind of no GPS scenario, pre-establishes the fine geomagnetic chart in maked an inspection tour region, carries out flying track conjecture using INS inertial navigation systems, provides robot present position；The earth magnetism characteristic quantity in current location earth's magnetic field is measured by the geomagnetic sensor entrained by itself；The geomagnetic data characteristic quantity of corresponding region is found out in accurate geomagnetic chart, match with geomagnetic data characteristic quantity measured by current location, the position that geomagnetic matching navigation provides is obtained, the state procedure navigated to inertial navigation system and geomagnetic matching using Extended Kalman filter is merged to obtain more accurate location information.Picture catching is carried out by the self-contained visual sensor of robot, the image collected is matched, position and the posture of inspecting robot are obtained, the state procedure with Unscented kalman filtering pair two compared with exact position merges, and obtains more accurate believable location information.
Description
Technical field
The invention belongs to localization for Mobile Robot navigation fields, and in particular to the complex region under a kind of no GPS scenario patrols
It is carried mainly for the space under Antagonistic Environment, the acquisition of time and location information depending on robot localization, navigation and time service method
Go out a kind of PNT (autonomous positioning, navigation and time service) method based under no GPS.
Background technology
At present using at most in terms of location and navigation technology, most ripe navigation mode has INS inertial navigations and GPS satellite
The method that navigation is combined.GPS satellite navigation has the advantages such as global, round-the-clock, long-time registration, but its signal is defeated
Go out discontinuous and be easily disturbed and block, under tree and high building is when blocking, and signal quality is deteriorated, and position inaccurate or can not obtain
Location information.INS inertial navigation systems are a kind of navigation modes of autonomous type, have very strong concealment and anti-interference ability,
Accurate positioning in short time.But MEMS-INS inertia systems due to device itself the characteristics of, gyroscope and acceleration are in respect of first
Beginning error, Random Drift Error etc., and cumulative errors are increasing at any time, long-time positioning accuracy is poor, can not obtain standard
True robot pose and location information.The method that GPS satellite navigation and the navigation of INS inertia systems are combined, when using GPS long
Between registration correct the deviation accumulation of inertial navigation system, can solve that GPS signal is discontinuous and inertial navigation system
The long-time deviation accumulation of system leads to problems such as positioning accuracy low.But it is weak in GPS signal or without GPS signal in the environment of can not be obtained
The posture and location information for taking robot, finally cannot be accurately positioned and navigate.
Invention content
Deficiency described in for the above-mentioned prior art, the present invention provide the complex region under a kind of no GPS scenario and make an inspection tour
Robot localization, navigation and time service method, it is weak or without GPS signal in GPS signal using intelligent security guard inspecting robot as carrier
Under environment, when GPS cannot be accurately positioned, using intelligent security guard inspecting robot of the present invention can precision navigation, complete make an inspection tour work(
The method of energy.
In order to solve the above technical problems, the technical solution adopted in the present invention is as follows：
The positioning of complex region inspecting robot, navigation and time service method under a kind of no GPS scenario, steps are as follows：
S1 builds accurate geomagnetic chart；
Specially：S1.1, the geomagnetic sensor that inspecting robot carries, which is treated, makes an inspection tour region progress geomagnetism detecting, and will
To earth magnetism characteristic quantity stored；
S1.2 treats and makes an inspection tour region progress mesh generation, obtains grid map；
S1.3 carries out the structure of the local geomagnetic chart based on Kriging interpolation to obtained earth magnetism characteristic quantity, after structure with net
Lattice map is matched, and accurate geomagnetic chart is obtained；
The structure of local geomagnetic chart is to use space auto-covariance optimal interpolation method, the main magnetic of the earth in local earth's magnetic field
3 ingredients of field, 3 main components of GEOMAGNETIC FIELD and disturbed magnetic field and Kriging technique assume there is one-to-one relationship, therefore
Kriging technique relatively accurately reflects the distribution space correlation properties rule in local earth's magnetic field, and the space for local earth's magnetic field is inserted
Value；
Detailed process is as follows：The expression formula of a certain magnetic field element value Z (X) at X is：
Z (X)=m (X)+ε ' (X)+ε "；
Wherein, m (X) is the determination function of geomagnetic main field structural component in describing Z (X)；ε ' (X) is description ground magnetic anomaly
Normal field random variation item related with spatial variations；ε " is the disturbed magnetic field unrelated with space；
Since the geomagnetic main field in some areas is gentle with spatial variations, multinomial is carried out by the geomagnetic data to actual measurement
Surface fitting and obtain, in local earth's magnetic field remove main field after, remaining magnetic field element value Z'(X) be earth magnetism regionality become
The sum of amount and disturbed magnetic field；
After structural component determines, remaining the change of divergence belongs to homogeneity variation, the difference between different location be only away from
From function,
Z (X) expression formula is changed to：
Z (X)=m (X)+λ (h)+ε "；
The estimation formula of wherein semivariance is：
Wherein, h is the distance between control point, is usually used in as lag coefficient；K is the quantity for the dominating pair of vertices for being apart h；
The measuring point of m estimation is numbered, be followed successively by 1,2 ..., m, difference point number is 0；For difference point X0It goes
Geomagnetic element estimated value after trend term；ωiFor the weight at i-th of control point；Zi' it is that i-th of measuring point removes the earth magnetism after trend term
Element value；
And the universal equation of Kriging technique estimation unknown point attribute value is：
When weight meets following formula, Z ' can be obtained0Minimum variance：
In formula：γ(hij) semivariance between point i and j in order to control；γ(hi0) point i and unknown point 0 in order to control semivariance；λ
For Lagrange coefficient；
Minimum variance S2For being reflected in the reliability of entire survey region interpolated value result；
S2, INS inertial navigation system carry out flying track conjecture；Geomagnetic sensor entrained by inspecting robot measures current location
Earth magnetism field data, and find out from geomagnetic chart the geomagnetic data of corresponding region.
Specially：S2.1, INS inertial navigation system carry out flying track conjecture；
Gyroscope in S2.1.1, INS inertial navigation system is used for forming a navigational coordinate system, makes the survey of accelerometer
Amount axis is stablized in navigational coordinate system and provides course and attitude angle；
S2.1.2 is passed through by the acceleration of accelerometer measures inspecting robot and obtains speed to the primary integral of time
Degree, integral formula are：
Vi=∫ aidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, aiFor the acceleration of the i-th moment inspecting robot, tiFor i-1
Moment is to the time interval between the i moment；
S2.1.3, speed obtain distance using the primary integral to the time, and integral formula is：
Li=∫ Vidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, LiFor the distance of the i-th moment inspecting robot, tiFor i-1 when
The time interval being carved between the i moment；
S2.1.4 carries out flying track conjecture by speed, distance to inspecting robot, obtains current location and predicts lower a period of time
The position at quarter, predictor formula are：
Xi=Li-1+Licosβi；
Wherein, βiFor attitude angle under current state；XiTo predict obtained subsequent time position, LiMachine is maked an inspection tour for the i-th moment
The distance of device people, Li-1For the distance of the (i-1)-th moment inspecting robot.
S2.2, geomagnetic sensor entrained by inspecting robot measure current location earth magnetism field data.
Inspecting robot obtains the earth magnetism characteristic quantity of current time current location by the geomagnetic sensor entrained by itself
(X, Y, Z), and be standardized, obtain standardization geomagnetic data；
The step of standardization is：
S2.2.1 seeks the coenvelope line Z of initial data Z (X)max(X), lower envelope line Zmin(X)；
The extreme point on initial data Z (X) is found out first, and then maximum point is inserted with cubic spline function curve
Value, fits the coenvelope line Z of initial data Z (X)max(X), lower envelope line Z is similarly obtainedmin(X)；
S2.2.2 seeks control point distance hi(X)；
hi(X)=Z (X)-δi(X)；
Wherein, δi(X) it is coenvelope line Zmax(X) and lower envelope line Zmin(X) mean value；
S2.2.3, by control point distance hi(X) regard original signal, repeat step S2.2.1-1~S2.2.1.2, obtain kth
The data h of secondary screeningk(X)；
S2.2.4 works as hk(X) meet the requirement of SD values, then hk(X) it is the first rank IMF, remembers Ci(x), from initial data Z (X)
In subtract Ci(x) residual error r is obtainedi(X)；
S2.2.5, as residual error ri(X) meet and be less than scheduled error or Ci(X) when being less than scheduled error, mode point is terminated
Solution preocess, the Z (X) of outputting standard, expression formula are：
Wherein, ri(X) it is acquired residual error in ith interpolation processing；Ci(X) it is to meet SD values in ith interpolation processing to want
The data asked；I is ith interpolation processing；N is interpolation processing total degree.
S2.3 finds out the geomagnetic data with inspecting robot current location earth magnetism field data corresponding region from geomagnetic chart.
S2.3.1, inertial navigation system provide the real time position of inspecting robot measurement time, the table in fine geomagnetic chart
It shows to carry out (Xi, Yi)；
S2.3.2 provides the time interval t at last moment to this moment by clocki, and provided and added by accelerometer
Speed ai, to time interval tiIt is integrated twice, obtains the distance L of two positionsI, formula is：
LI=∫ ∫ aidtidti；
S2.3.3, with last moment matching position (XI,YI) it is origin, and with current location and last moment matching position
Distance LIFor radius, circular arc is done, obtains the geomagnetic matching region at current location moment, and the acquisition from fine geomagnetic chart
Earth magnetism characteristic quantity data with region.
The earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization is carried out earth magnetism by S3
Match, and the state procedure navigated to inertial navigation system and geomagnetic matching is merged to obtain more accurate location information.
Specially：S3.1, by the earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization
Carry out geomagnetic matching；
Matching uses the sectoring method based on ICP algorithm, and steps are as follows：
1), according to traditional ICP flows, ICP algorithm is made to converge to locally best matchingRotation group controls e=0；
2) rotation of e groups, rotation step number τ=0, rotary step t, threshold rotating value ξ, are carried out：
1. withA fixed endpoint be origin, rotate (τ+1) × t degree to the convergence direction of ICP；
2. after rotation, optimal transformation is sought using Quaternion Method
3. enabling τ=τ+1；Repeat step 1. with step 2., until (τ+1) × t >=ξ；
4. after every group of rotation,In selected most preferably according to measurement error quadratic sum minimum principle
Matching
5. replacing rotation endpoint, e=e+1 is enabled；Repeat step 1. -4., until meeting end condition.Obtain best match
Point, the stopping criterion for iteration are that iterations or every group of optimal value variation rotated are sufficiently small.
S3.2, the state procedure navigated to inertial navigation system and geomagnetic matching with Extended Kalman filter merge
Obtain more being accurately positioned for current location；
The specific steps are：
S3.2.1, init state vector and variance：
S3.2.2 establishes 13 dimension state equations：
Wherein,For the state variable of system t moment, F (t) is t moment state-transition matrix, and G (t) makes an uproar for t moment
Sound transfer matrix, W (t) are system noise；
S3.2.3 establishes predictive equation；
Assuming that having obtained the state estimation at this moment at the K moment, it is denoted asPass through state estimations prediction equation K+1
The estimated value at moment；
S3.2.3.1, the one-step prediction equation for establishing quantity of state are：
Wherein, ωiFor with XiCorresponding weights；
S3.2.3.2 establishes the one-step prediction error of quantity of state：
Xi,k+1/kFor state-transition matrix；
S3.2.3.3 establishes the one-step prediction of observed quantity：
Zi,k+1/kTo observe transfer matrix；
S3.2.3.4 establishes the one-step prediction error of observed quantity：
To observe covariance matrix；
S3.2.4, renewal equation；
S3.2.4.1 calculates filtering gain matrix：
WK+1=PK+1/K(PK+1/K)-1；
S3.2.4.2 calculates filtering estimated value：
S3.2.4.3 calculates filtering error variance matrix：
S4 carries out vision guided navigation positioning, by the feature extracting and matching of image between station, it can be achieved that image overlapping region
Sub-pix matches, and position and the posture of inspecting robot are obtained using bundle adjustment using matched result.
Using the binocular navigation camera of the advance Accurate Calibration mounted on inspecting robot, sequence chart is shot in different location
Picture is settled accounts by images match and the three-dimensional relationship of image characteristic point, obtains position and the posture of inspecting robot；
S4.1 is imaged with navigation camera in two matching positions, there is certain degree of overlapping between the image of two cameras respectively,
Common characteristic point first is selected from adjacent two positions left image overlapping region, is then carried out between the stereo-picture of same position
Accurate matching uses bundle adjustment that the relative positioning between two positions is calculated by the characteristic point matched.
S4.2 carries out feature extracting and matching in the left camera image of different location shooting, extracts matching characteristic point, and
Characteristic point of the same name is extracted in right camera image according to left camera image characteristic point, then using these characteristic points as observation point,
STFT matchings are carried out in adjacent two positions overlapping region, obtain the characteristic point matched.
S4.3 carries out correlation coefficient matching method, in matched result between a upper position and the stereo pairs of current location
On the basis of carry out Least squares matching, realize sub-pix images match, and multiple strategy is carried out to matching result and is detected, reject
The rough error matched.
The seat for realizing matched characteristic point camera coordinates system under current time is calculated by three-dimensional imaging model by S4.4
Mark will match point coordinates and be transformed into robot at this further according to the conversion parameter between camera coordinates system and robot coordinate system
Similarly adjacent position S is calculated in coordinate under the robot coordinate system at quartern+1Lower match point is sat in current location inspecting robot
The lower coordinate of mark system.
S4.5, by SnLower match point and Sn+1Lower match point coordinate under the inspecting robot coordinate system of current location, obtains phase
The translation parameters between inspecting robot coordinate system is set at ortho position, and rotation parameter further obtains Sn+1Inspecting robot coordinate under position
The pose under global coordinate system is calculated in translation parameters between system and global coordinate system, rotation parameter.
S4.5.1, the overall situation that the bodywork reference frame that inspecting robot sets station location at first is investigated as inspecting robot
Coordinate system, by visual sensor installation parameter, the body coordinate system and camera coordinates of inspecting robot when calculating first position
Conversion parameter between system, i.e. translation parameters (X0,Y0,Z0) and rotation parameter (RX,RY,RZ)。
S4.5.2 obtains the coordinate under the current position vision system coordinate system by accordingly matching, then by Current vision system
Matching point coordinates is transformed into coordinate under the body coordinate system of inspecting robot in the position by the corner and pitch angle of systemSimilarly, adjacent position S is obtainedn+1Under match point coordinate under body coordinate system in the position
S4.5.3 is obtained by calculation by two adjacent position coordinates between two adjacent position inspecting robot body coordinate systems
Translation parametersAnd rotation parameterRecursion obtains adjacent position S againn+1It is sat with the overall situation
Translation parameters between mark systemAnd rotation parameter
S4.5.4 is exactly current location inspecting robot in global coordinate system relative to the translation parameters under global coordinate system
Under position, rotation parameter calculates the posture under global coordinate system.
S5, by the tour for being more accurately positioned with being obtained in step S4 for the inspecting robot current location that step S3 is obtained
The position of robot carries out Unscented kalman filtering fusion, obtains more accurate believable location information.
The accurate location information that step S5 is obtained is fed back to inertial navigation system and carries out data update, to repair by S6
The accumulated error of positive inertial navigation system.
The present invention pre-establishes the fine geomagnetic chart in maked an inspection tour region, and flying track conjecture is carried out using INS inertial navigation systems,
Provide robot present position；The earth magnetism feature in current location earth's magnetic field is measured by the geomagnetic sensor entrained by itself
Amount；The geomagnetic data characteristic quantity that corresponding region is found out in accurate geomagnetic chart, with geomagnetic data characteristic quantity measured by current location
Match, the position that geomagnetic matching navigation provides is obtained, using Extended Kalman filter to inertial navigation system and geomagnetic matching
The state procedure of navigation is merged to obtain more accurate location information.Figure is carried out by the self-contained visual sensor of robot
As capturing, the image collected is matched, position and the posture of inspecting robot are obtained, with Unscented kalman filtering pair
Two state procedures compared with exact position are merged, and more accurate believable location information is obtained.By accurate location information
It feeds back to inertial navigation system and carries out data update, to correct the accumulated error of inertial navigation system.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Fig. 2 is vision matching flow chart of the present invention.
Specific implementation mode
As shown in Figure 1, the positioning of complex region inspecting robot, navigation and time service method under a kind of no GPS scenario, step
It is as follows：
S1 builds accurate geomagnetic chart；
Specially：S1.1, the geomagnetic sensor that inspecting robot carries, which is treated, makes an inspection tour region progress geomagnetism detecting, and will
To earth magnetism characteristic quantity stored；
S1.2 treats and makes an inspection tour region progress mesh generation, obtains grid map；
S1.3 carries out the structure of the local geomagnetic chart based on Kriging interpolation to obtained earth magnetism characteristic quantity, after structure with net
Lattice map is matched, and accurate geomagnetic chart is obtained；
The structure of local geomagnetic chart is to use space auto-covariance optimal interpolation method, the main magnetic of the earth in local earth's magnetic field
3 ingredients of field, 3 main components of GEOMAGNETIC FIELD and disturbed magnetic field and Kriging technique assume there is one-to-one relationship, therefore
Kriging technique relatively accurately reflects the distribution space correlation properties rule in local earth's magnetic field, and the space for local earth's magnetic field is inserted
Value；
Detailed process is as follows：The expression formula of a certain magnetic field element value Z (X) at X is：
Z (X)=m (X)+ε ' (X)+ε "；
Wherein, m (X) is the determination function of geomagnetic main field structural component in describing Z (X)；ε ' (X) is description ground magnetic anomaly
Normal field random variation item related with spatial variations；ε " is the disturbed magnetic field unrelated with space；
Since the geomagnetic main field in some areas is gentle with spatial variations, multinomial is carried out by the geomagnetic data to actual measurement
Surface fitting and obtain, in local earth's magnetic field remove main field after, remaining magnetic field element value Z'(X) be earth magnetism regionality become
The sum of amount and disturbed magnetic field；
After structural component determines, remaining the change of divergence belongs to homogeneity variation, the difference between different location be only away from
From function,
Z (X) expression formula is changed to：
Z (X)=m (X)+λ (h)+ε "；
The estimation formula of wherein semivariance is：
Wherein, h is the distance between control point, is usually used in as lag coefficient；K is the quantity for the dominating pair of vertices for being apart h；
The measuring point of m estimation is numbered, be followed successively by 1,2 ..., m, difference point number is 0；For difference point X0It goes
Geomagnetic element estimated value after trend term；ωiFor the weight at i-th of control point；Zi' it is that i-th of measuring point removes the earth magnetism after trend term
Element value；
And the universal equation of Kriging technique estimation unknown point attribute value is：
When weight meets following formula, Z ' can be obtained0Minimum variance：
In formula：γ(hij) semivariance between point i and j in order to control；γ(hi0) point i and unknown point 0 in order to control semivariance；λ
For Lagrange coefficient；
Minimum variance S2For being reflected in the reliability of entire survey region interpolated value result；
S2, INS inertial navigation system carry out flying track conjecture；Geomagnetic sensor entrained by inspecting robot measures current location
Earth magnetism field data, and find out from geomagnetic chart the geomagnetic data of corresponding region.
Specially：S2.1, INS inertial navigation system carry out flying track conjecture；
Gyroscope in S2.1.1, INS inertial navigation system is used for forming a navigational coordinate system, makes the survey of accelerometer
Amount axis is stablized in navigational coordinate system and provides course and attitude angle；
S2.1.2 is passed through by the acceleration of accelerometer measures inspecting robot and obtains speed to the primary integral of time
Degree, integral formula are：
Vi=∫ aidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, aiFor the acceleration of the i-th moment inspecting robot, tiFor i-1
Moment is to the time interval between the i moment；
S2.1.3, speed obtain distance using the primary integral to the time, and integral formula is：
Li=∫ Vidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, LiFor the distance of the i-th moment inspecting robot, tiFor i-1 when
The time interval being carved between the i moment；
S2.1.4 carries out flying track conjecture by speed, distance to inspecting robot, obtains current location and predicts lower a period of time
The position at quarter, predictor formula are：
Xi=Li-1+Licosβi；
Wherein, βiFor attitude angle under current state；XiTo predict obtained subsequent time position, LiMachine is maked an inspection tour for the i-th moment
The distance of device people, Li-1For the distance of the (i-1)-th moment inspecting robot.
S2.2, geomagnetic sensor entrained by inspecting robot measure current location earth magnetism field data.
Inspecting robot obtains the earth magnetism characteristic quantity of current time current location by the geomagnetic sensor entrained by itself
(X, Y, Z), and be standardized, obtain standardization geomagnetic data；
The step of standardization is：
S2.2.1 seeks the coenvelope line Z of initial data Z (X)max(X), lower envelope line Zmin(X)；
The extreme point on initial data Z (X) is found out first, and then maximum point is inserted with cubic spline function curve
Value, fits the coenvelope line Z of initial data Z (X)max(X), lower envelope line Z is similarly obtainedmin(X)；
S2.2.2 seeks control point distance hi(X)；
hi(X)=Z (X)-δi(X)；
Wherein, δi(X) it is coenvelope line Zmax(X) and lower envelope line Zmin(X) mean value；
S2.2.3, by control point distance hi(X) regard original signal, repeat step S2.2.1-1~S2.2.1.2, obtain kth
The data h of secondary screeningk(X)；
S2.2.4 works as hk(X) meet the requirement of SD values, then hk(X) it is the first rank IMF, remembers Ci(x), from initial data Z (X)
In subtract Ci(x) residual error r is obtainedi(X)；
S2.2.5, as residual error ri(X) meet and be less than scheduled error or Ci(X) when being less than scheduled error, mode point is terminated
Solution preocess, the Z (X) of outputting standard, expression formula are：
Wherein, ri(X) it is acquired residual error in ith interpolation processing；Ci(X) it is to meet SD values in ith interpolation processing to want
The data asked；I is ith interpolation processing；N is interpolation processing total degree.
S2.3 finds out the geomagnetic data with inspecting robot current location earth magnetism field data corresponding region from geomagnetic chart.
S2.3.1, inertial navigation system provide the real time position of inspecting robot measurement time, the table in fine geomagnetic chart
It shows to carry out (Xi, Yi)；
S2.3.2 provides the time interval t at last moment to this moment by clocki, and provided and added by accelerometer
Speed ai, to time interval tiIt is integrated twice, obtains the distance L of two positionsI, formula is：
LI=∫ ∫ aidtidti；
S2.3.3, with last moment matching position (XI,YI) it is origin, and with current location and last moment matching position
Distance LIFor radius, circular arc is done, obtains the geomagnetic matching region at current location moment, and the acquisition from fine geomagnetic chart
Earth magnetism characteristic quantity data with region.
The earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization is carried out earth magnetism by S3
Match, and the state procedure navigated to inertial navigation system and geomagnetic matching is merged to obtain more accurate location information.
Specially：S3.1, by the earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization
Carry out geomagnetic matching；
Matching uses the sectoring method based on ICP algorithm, and steps are as follows：
1), according to traditional ICP flows, ICP algorithm is made to converge to locally best matchingRotation group controls e=0；
2) rotation of e groups, rotation step number τ=0, rotary step t, threshold rotating value ξ, are carried out：
1. withA fixed endpoint be origin, rotate (τ+1) × t degree to the convergence direction of ICP；
2. after rotation, optimal transformation is sought using Quaternion Method
3. enabling τ=τ+1；Repeat step 1. with step 2., until (τ+1) × t >=ξ；
4. after every group of rotation,In selected most preferably according to measurement error quadratic sum minimum principle
Matching
5. replacing rotation endpoint, e=e+1 is enabled；Repeat step 1. -4., until meeting end condition.Obtain best match
Point, the stopping criterion for iteration are that iterations or every group of optimal value variation rotated are sufficiently small.
S3.2, the state procedure navigated to inertial navigation system and geomagnetic matching with Extended Kalman filter merge
Obtain more being accurately positioned for current location；
The specific steps are：
S3.2.1, init state vector and variance：
S3.2.2 establishes 13 dimension state equations：
Wherein,For the state variable of system t moment, F (t) is t moment state-transition matrix, and G (t) makes an uproar for t moment
Sound transfer matrix, W (t) are system noise；
S3.2.3 establishes predictive equation；
Assuming that having obtained the state estimation at this moment at the K moment, it is denoted asPass through state estimations prediction equation K+1
The estimated value at moment；
S3.2.3.1, the one-step prediction equation for establishing quantity of state are：
Wherein, ωiFor with XiCorresponding weights；
S3.2.3.2 establishes the one-step prediction error of quantity of state：
Xi,k+1/kFor state-transition matrix；
S3.2.3.3 establishes the one-step prediction of observed quantity：
Zi,k+1/kTo observe transfer matrix；
S3.2.3.4 establishes the one-step prediction error of observed quantity：
To observe covariance matrix；
S3.2.4, renewal equation；
S3.2.4.1 calculates filtering gain matrix：
WK+1=PK+1/K(PK+1/K)-1；
S3.2.4.2 calculates filtering estimated value：
S3.2.4.3 calculates filtering error variance matrix：
S4 carries out vision guided navigation positioning, by the feature extracting and matching of image between station, it can be achieved that image overlapping region
Sub-pix matches, and position and the posture of inspecting robot are obtained using bundle adjustment using matched result.
Using the binocular navigation camera of the advance Accurate Calibration mounted on inspecting robot, sequence chart is shot in different location
Picture is settled accounts by images match and the three-dimensional relationship of image characteristic point, obtains position and the posture of inspecting robot；
S4.1 is imaged with navigation camera in two matching positions, there is certain degree of overlapping between the image of two cameras respectively,
Common characteristic point first is selected from adjacent two positions left image overlapping region, is then carried out between the stereo-picture of same position
Accurate matching uses bundle adjustment that the relative positioning between two positions is calculated by the characteristic point matched.
S4.2 carries out feature extracting and matching in the left camera image of different location shooting, extracts matching characteristic point, and
Characteristic point of the same name is extracted in right camera image according to left camera image characteristic point, then using these characteristic points as observation point,
STFT matchings are carried out in adjacent two positions overlapping region, obtain the characteristic point matched.
S4.3 carries out correlation coefficient matching method, in matched result between a upper position and the stereo pairs of current location
On the basis of carry out Least squares matching, realize sub-pix images match, and multiple strategy is carried out to matching result and is detected, reject
The rough error matched.
The seat for realizing matched characteristic point camera coordinates system under current time is calculated by three-dimensional imaging model by S4.4
Mark will match point coordinates and be transformed into robot at this further according to the conversion parameter between camera coordinates system and robot coordinate system
Similarly adjacent position S is calculated in coordinate under the robot coordinate system at quartern+1Lower match point is sat in current location inspecting robot
The lower coordinate of mark system.
S4.5, by SnLower match point and Sn+1Lower match point coordinate under the inspecting robot coordinate system of current location, obtains phase
The translation parameters between inspecting robot coordinate system is set at ortho position, and rotation parameter further obtains Sn+1Inspecting robot coordinate under position
The pose under global coordinate system is calculated in translation parameters between system and global coordinate system, rotation parameter.
S4.5.1, the overall situation that the bodywork reference frame that inspecting robot sets station location at first is investigated as inspecting robot
Coordinate system, by visual sensor installation parameter, the body coordinate system and camera coordinates of inspecting robot when calculating first position
Conversion parameter between system, i.e. translation parameters (X0,Y0,Z0) and rotation parameter (RX,RY,RZ)。
S4.5.2 obtains the coordinate under the current position vision system coordinate system by accordingly matching, then by Current vision system
Matching point coordinates is transformed into coordinate under the body coordinate system of inspecting robot in the position by the corner and pitch angle of systemSimilarly, adjacent position S is obtainedn+1Under match point coordinate under body coordinate system in the position
S4.5.3 is obtained by calculation by two adjacent position coordinates between two adjacent position inspecting robot body coordinate systems
Translation parametersAnd rotation parameterRecursion obtains adjacent position S againn+1It is sat with the overall situation
Translation parameters between mark systemAnd rotation parameter
S4.5.4 is exactly current location inspecting robot in global coordinate system relative to the translation parameters under global coordinate system
Under position, rotation parameter calculates the posture under global coordinate system.
S5, by the tour for being more accurately positioned with being obtained in step S4 for the inspecting robot current location that step S3 is obtained
The position of robot carries out Unscented kalman filtering fusion, obtains more accurate believable location information.
The position vector obtained after earth-magnetism navigation system is merged with INS inertial navigation systems will be regarded as state vector
Feel that navigator fix obtains location information as observed quantity, these deterministic sampling points are carried out respectively corresponding to the non-of state equation
Final state variable is estimated in linear transformation after weighted calculating.
It is as follows：
S5.1 builds state equation：
Xk=f (Xk-1)+Wk-1；
S5.2 builds observational equation：
Zk=h (Xk)+Vk；
S5.3 enables the mean value of state vector X and variance be expressed asAnd Px, convert, obtain according to UTThe 2n+1 of surrounding
A sigma points：
Wherein, λ=α2(n+k)-n；α be adjust sigma points andDistance, it is guarantor usually to take the decimal more than 0, k
Poor battle array positive semidefinite and the minor parameter being arranged；
S5.4 defines XiCorresponding weights：
S5.5 carries out the nonlinear transformation of sampled point by nonlinear state function f after each sample：
S5.6 is weighted estimation to the sampled point after transformation：
S5.7 establishes status predication equation：
S5.8 establishes measurement equation：
The accurate location information that step S5 is obtained is fed back to inertial navigation system and carries out data update, to repair by S6
The accumulated error of positive inertial navigation system, to complete the PNT at no GPS.
Described above is only presently preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention god.
Claims (10)
1. the complex region inspecting robot location navigation under a kind of no GPS scenario and time service method, which is characterized in that step is such as
Under：S1 builds accurate geomagnetic chart；
S2, INS inertial navigation system carry out flying track conjecture；Geomagnetic sensor entrained by inspecting robot measures current location earth magnetism
Field data；And the geomagnetic data of corresponding region is found out from geomagnetic chart；
The earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization is carried out geomagnetic matching by S3,
And it is merged to obtain with the state procedure that Extended Kalman filter navigates to inertial navigation system and geomagnetic matching more smart
True location information；
S4 carries out vision guided navigation positioning, obtains position and the posture of inspecting robot；
Using the binocular navigation camera of the advance Accurate Calibration mounted on inspecting robot, sequence image is shot in different location,
It is settled accounts by images match and the three-dimensional relationship of image characteristic point, obtains position and the posture of inspecting robot；
S5, by the tour machine for being more accurately positioned with being obtained in step S4 for the inspecting robot current location that step S3 is obtained
The position of people carries out Unscented kalman filtering fusion, obtains more accurate believable location information；
The accurate location information that step S5 is obtained is fed back to inertial navigation system and carries out data update by S6, used to correct
The accumulated error of property navigation system.
2. the complex region inspecting robot location navigation under no GPS scenario according to claim 1 and time service method,
It is characterized in that, in step sl, the specific steps are：
S1.1, the geomagnetic sensor that inspecting robot carries, which is treated, makes an inspection tour region progress geomagnetism detecting, and obtained earth magnetism is special
Sign amount is stored；
S1.2 treats and makes an inspection tour region progress mesh generation, obtains grid map；
S1.3 carries out the local geomagnetic chart based on Kriging interpolation to obtained earth magnetism characteristic quantity and builds, after structure with grid
Figure is matched, and accurate geomagnetic chart is obtained；
The structure of local geomagnetic chart is to use space auto-covariance optimal interpolation method, geomagnetic main field in local earth's magnetic field,
3 ingredients of 3 main components of GEOMAGNETIC FIELD and disturbed magnetic field and Kriging technique assume have one-to-one relationship, therefore gram in
Golden method relatively accurately reflects the distribution space correlation properties rule in local earth's magnetic field, the space interpolation for local earth's magnetic field；
Detailed process is as follows：The expression formula of a certain magnetic field element value Z (X) at X is：
Z (X)=m (X)+ε ' (X)+ε "；
Wherein, m (X) is the determination function of geomagnetic main field structural component in describing Z (X)；ε ' (X) is description GEOMAGNETIC FIELD
Random variation item related with spatial variations；ε " is the disturbed magnetic field unrelated with space；
Since the geomagnetic main field in some areas is gentle with spatial variations, polynomial surface is carried out by the geomagnetic data to actual measurement
Fitting and obtains, in local earth's magnetic field remove main field after, remaining magnetic field element value Z'(X) be earth magnetism areal variable with
The sum of disturbed magnetic field；
After structural component determines, remaining the change of divergence belongs to homogeneity variation, and the difference between different location is only distance
Function,
Z (X) expression formula is changed to：
Z (X)=m (X)+λ (h)+ε "；
The estimation formula of wherein semivariance is：
Wherein, h is the distance between control point, is usually used in as lag coefficient；K is the quantity for the dominating pair of vertices for being apart h；
The measuring point of m estimation is numbered, be followed successively by 1,2 ..., m, difference point number is 0；For difference point X0Go trend
Geomagnetic element estimated value after；ωiFor the weight at i-th of control point；Zi' it is that i-th of measuring point removes the geomagnetic element after trend term
Value；
And the universal equation of Kriging technique estimation unknown point attribute value is：
When weight meets following formula, Z can be obtained0' minimum variance：
In formula：γ(hij) semivariance between point i and j in order to control；γ(hi0) point i and unknown point 0 in order to control semivariance；λ is to draw
Ge Lang coefficients；
Minimum variance S2For being reflected in the reliability of entire survey region interpolated value result；
3. the complex region inspecting robot location navigation under no GPS scenario according to claim 1 and time service method,
It is characterized in that, in step s 2, the specific steps are：
S2.1, INS inertial navigation system carry out flying track conjecture；
S2.2, geomagnetic sensor entrained by inspecting robot measure current location earth magnetism field data；
Inspecting robot by the geomagnetic sensor entrained by itself, obtain current time current location earth magnetism characteristic quantity (X,
Y, Z), and be standardized, obtain standardization geomagnetic data；
S2.3 finds out the geomagnetic data with inspecting robot current location earth magnetism field data corresponding region from geomagnetic chart.
4. the complex region inspecting robot location navigation under no GPS scenario according to claim 3 and time service method,
It is characterized in that, in step S2.1, the specific steps are：
Gyroscope in S2.1.1, INS inertial navigation system is used for forming a navigational coordinate system, makes the measurement axis of accelerometer
Stablize in navigational coordinate system and provides course and attitude angle；
S2.1.2 is passed through by the acceleration of accelerometer measures inspecting robot and obtains speed to the primary integral of time, product
Point formula is：
Vi=∫ aidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, aiFor the acceleration of the i-th moment inspecting robot, tiFor the i-1 moment
Time interval between the i moment；
S2.1.3, speed obtain distance using the primary integral to the time, and integral formula is：
Li=∫ Vidti；
Wherein, ViFor the speed of the i-th moment inspecting robot, LiFor the distance of the i-th moment inspecting robot, tiIt is arrived for the i-1 moment
Time interval between the i moment；
S2.1.4 carries out flying track conjecture to inspecting robot by speed, distance, obtains current location and predict subsequent time
Position, predictor formula are：
Xi=Li-1+Licosβi；
Wherein, βiFor attitude angle under current state；XiFor i moment predicted positions, LiFor the distance of the i-th moment inspecting robot,
Li-1For the distance of the (i-1)-th moment inspecting robot.
5. the complex region inspecting robot location navigation under no GPS scenario according to claim 3 and time service method,
It is characterized in that, in step S2.2, the step of standardization is：
S2.2.1 seeks the coenvelope line Z of initial data Z (X)max(X), lower envelope line Zmin(X)；
The extreme point on initial data Z (X) is found out first, then uses cubic spline function curve to maximum point into row interpolation,
Fit the coenvelope line Z of initial data Z (X)max(X), lower envelope line Z is similarly obtainedmin(X)；
S2.2.2 seeks control point distance hi(X)；
hi(X)=Z (X)-δi(X)；
Wherein, δi(X) it is coenvelope line Zmax(X) and lower envelope line Zmin(X) mean value；
S2.2.3, by control point distance hi(X) regard original signal, repeat step S2.2.1-1~S2.2.1.2, obtain kth time sieve
The data h of choosingk(X)；
S2.2.4 works as hk(X) meet the requirement of SD values, then hk(X) it is the first rank IMF, remembers Ci(x), subtract from initial data Z (X)
Remove Ci(x) residual error r is obtainedi(X)；
S2.2.5, as residual error ri(X) meet and be less than scheduled error or Ci(X) when being less than scheduled error, mode decomposition mistake is terminated
Journey, the Z (X) of outputting standard, expression formula are：
Wherein, ri(X) it is acquired residual error in ith interpolation processing；Ci(X) it is to meet the requirement of SD values in ith interpolation processing
Data；I is ith interpolation processing；N is interpolation processing total degree.
6. the complex region inspecting robot location navigation under no GPS scenario according to claim 3 and time service method,
It is characterized in that, in step S2.3, the specific steps are：
S2.3.1, inertial navigation system provide the real time position of inspecting robot measurement time, are represented in fine geomagnetic chart
Carry out (Xi, Yi)；
S2.3.2 provides the time interval t at last moment to this moment by clocki, and acceleration is provided by accelerometer
ai, to time interval tiIt is integrated twice, obtains the distance L of two positionsI, formula is：
LI=∫ ∫ aidtidti；
S2.3.3, with last moment matching position (XI,YI) be origin, and with current location and last moment matching position away from
From LIFor radius, circular arc is done, obtains the geomagnetic matching region at current location moment, and matching area is obtained from fine geomagnetic chart
Earth magnetism characteristic quantity data.
7. the complex region inspecting robot location navigation under no GPS scenario according to claim 1 and time service method,
It is characterized in that, in step s3, the specific steps are：
The earth magnetism characteristic quantity data of current location earth magnetism field data and matching area after standardization is carried out earth magnetism by S3.1
Match；
Matching uses the sectoring method based on ICP algorithm, and steps are as follows：
1), according to traditional ICP flows, ICP algorithm is made to converge to locally best matchingRotation group controls e=0；
2) rotation of e groups, rotation step number τ=0, rotary step t, threshold rotating value ξ, are carried out；
1. withA fixed endpoint be origin, rotate (τ+1) × t degree to the convergence direction of ICP；
2. after rotation, optimal transformation is sought using Quaternion Method
3. enabling τ=τ+1；Repeat step 1. with step 2., until (τ+1) × t >=ξ；
4. after every group of rotation,According to measurement error quadratic sum minimum principle select best match
5. replacing rotation endpoint, e=e+1 is enabled；Repeat step 1. -4., until meeting end condition；Optimal match point is obtained, institute
It is that iterations or every group of optimal value variation rotated are sufficiently small to state stopping criterion for iteration；
S3.2, the state procedure navigated to inertial navigation system and geomagnetic matching with Extended Kalman filter are merged to obtain
Current location is more accurately positioned.
8. the complex region inspecting robot location navigation under no GPS scenario according to claim 7 and time service method,
It is characterized in that, in step S3.2, the specific steps are：The specific steps are：
S3.2.1, init state vector and variance：
S3.2.2 establishes 13 dimension state equations：
Wherein,For the state variable of system t moment, F (t) is t moment state-transition matrix, and G (t) shifts for t moment noise
Matrix, W (t) are system noise；
S3.2.3 establishes predictive equation；
Assuming that having obtained the state estimation at this moment at the K moment, it is denoted asPass through the state estimations prediction equation K+1 moment
Estimated value；
S3.2.3.1, the one-step prediction equation for establishing quantity of state are：
Wherein, ωiFor with XiCorresponding weights；
S3.2.3.2 establishes the one-step prediction error of quantity of state：
Xi,k+1/kFor state-transition matrix；
S3.2.3.3 establishes the one-step prediction of observed quantity：
Zi,k+1/kTo observe transfer matrix；
S3.2.3.4 establishes the one-step prediction error of observed quantity：
To observe covariance matrix；
S3.2.4, renewal equation；
S3.2.4.1 calculates filtering gain matrix：
WK+1=PK+1/K(PK+1/K)-1；
S3.2.4.2 calculates filtering estimated value：
S3.2.4.3 calculates filtering error variance matrix：
9. the complex region inspecting robot location navigation under no GPS scenario according to claim 1 and time service method,
It is characterized in that, in step s 4, the specific steps are：
S4.1 is imaged with navigation camera in two matching positions, there is certain degree of overlapping between the image of two cameras respectively, first from
Adjacent two positions left image selects common characteristic point in overlapping region, is then carried out between the stereo-picture of same position accurate
Matching, the relative positioning between two positions is calculated using bundle adjustment by the characteristic point that matches；
S4.2 carries out feature extracting and matching in the left camera image of different location shooting, extracts matching characteristic point, and according to
Left camera image characteristic point extracts characteristic point of the same name in right camera image, then using these characteristic points as observation point, in phase
Adjacent two positions overlapping region carries out STFT matchings, obtains the characteristic point matched；
S4.3 carries out correlation coefficient matching method between a upper position and the stereo pairs of current location, on matched result basis
Upper carry out Least squares matching realizes sub-pix images match, and carries out multiple strategy to matching result and detect, and rejects matched
Rough error；
The coordinate for realizing matched characteristic point camera coordinates system under current time is calculated by three-dimensional imaging model by S4.4,
Further according to the conversion parameter between camera coordinates system and robot coordinate system, matching point coordinates is transformed into robot at the moment
Similarly adjacent position S is calculated in coordinate under robot coordinate systemn+1Lower match point is in current location inspecting robot coordinate system
Lower coordinate；
S4.5, by position SnLower match point and Sn+1Lower match point coordinate under the inspecting robot coordinate system of current location obtains adjacent
Translation parameters between the inspecting robot coordinate system of position, rotation parameter further obtain Sn+1Inspecting robot coordinate system under position
The pose under global coordinate system is calculated in translation parameters between global coordinate system, rotation parameter.
10. the complex region inspecting robot location navigation under no GPS scenario according to claim 9 and time service method,
It is characterized in that, in step S4.5, the specific steps are：
S4.5.1, the world coordinates that the bodywork reference frame that inspecting robot sets station location at first is investigated as inspecting robot
System, by visual sensor installation parameter, when calculating first position between the body coordinate system and camera coordinates system of inspecting robot
Conversion parameter, i.e. translation parameters (X0,Y0,Z0) and rotation parameter (RX,RY,RZ)；
S4.5.2 obtains the coordinate under the current position vision system coordinate system by accordingly matching, then by Current vision system
Matching point coordinates is transformed into coordinate under the body coordinate system of inspecting robot in the position by corner and pitch angleSimilarly, adjacent position S is obtainedn+1Under match point coordinate under body coordinate system in the position
The translation between two adjacent position inspecting robot body coordinate systems is obtained by calculation by two adjacent position coordinates in S4.5.3
ParameterAnd rotation parameterRecursion obtains adjacent position S againn+1With global coordinate system
Between translation parametersAnd rotation parameter
S4.5.4 is exactly current location inspecting robot under global coordinate system relative to the translation parameters under global coordinate system
Position, rotation parameter calculate the posture under global coordinate system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810305986.1A CN108592914A (en) | 2018-04-08 | 2018-04-08 | The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810305986.1A CN108592914A (en) | 2018-04-08 | 2018-04-08 | The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108592914A true CN108592914A (en) | 2018-09-28 |
Family
ID=63621142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810305986.1A Pending CN108592914A (en) | 2018-04-08 | 2018-04-08 | The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108592914A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110455298A (en) * | 2019-08-14 | 2019-11-15 | 灵动科技（北京）有限公司 | Vehicle localization method and positioning system |
| CN110648398A (en) * | 2019-08-07 | 2020-01-03 | 武汉九州位讯科技有限公司 | Real-time ortho image generation method and system based on unmanned aerial vehicle aerial data |
| CN112013842A (en) * | 2020-08-29 | 2020-12-01 | 桂林电子科技大学 | Multi-mode indoor positioning method based on image geomagnetic field and inertial sensor |
| CN112147995A (en) * | 2019-06-28 | 2020-12-29 | 深圳市创客工场科技有限公司 | Robot motion control method and device, robot and storage medium |
| CN112560974A (en) * | 2020-12-22 | 2021-03-26 | 清华大学 | Information fusion and vehicle information acquisition method and device |
| CN113739784A (en) * | 2020-05-27 | 2021-12-03 | 华为技术有限公司 | Positioning method, user equipment, storage medium and electronic equipment |
| CN113873442A (en) * | 2021-09-08 | 2021-12-31 | 宁波大榭招商国际码头有限公司 | External hub card positioning method |
| CN114111808A (en) * | 2021-11-30 | 2022-03-01 | 上汽通用五菱汽车股份有限公司 | Positioning method, system and device of unmanned vehicle and readable storage medium |
| CN114117113A (en) * | 2022-01-28 | 2022-03-01 | 杭州宏景智驾科技有限公司 | Multi-feature-point motor vehicle positioning method and device, electronic equipment and storage medium |
| CN114136309A (en) * | 2021-11-12 | 2022-03-04 | 上汽通用五菱汽车股份有限公司 | Positioning method, system and device of unmanned logistics vehicle and storage medium |
| CN116698018A (en) * | 2023-08-08 | 2023-09-05 | 山西戴德测控技术股份有限公司 | Navigation positioning auxiliary device and coal mine tunnel navigation positioning system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927739A (en) * | 2014-01-10 | 2014-07-16 | 北京航天飞行控制中心 | Patroller positioning method based on spliced images |
| CN105222772A (en) * | 2015-09-17 | 2016-01-06 | 泉州装备制造研究所 | A kind of high-precision motion track detection system based on Multi-source Information Fusion |
| CN105424030A (en) * | 2015-11-24 | 2016-03-23 | 东南大学 | Fusion navigation device and method based on wireless fingerprints and MEMS sensor |
| US20160360501A1 (en) * | 2014-12-18 | 2016-12-08 | Alibaba Group Holding Limited | Method and apparatus of positioning mobile terminal based on geomagnetism |
| CN106767828A (en) * | 2016-12-29 | 2017-05-31 | 南京邮电大学 | A kind of mobile phone indoor positioning solution |
| CN106780699A (en) * | 2017-01-09 | 2017-05-31 | 东南大学 | A kind of vision SLAM methods aided in based on SINS/GPS and odometer |
| CN107144281A (en) * | 2017-06-30 | 2017-09-08 | 飞智控（天津）科技有限公司 | Unmanned plane indoor locating system and localization method based on cooperative target and monocular vision |
| CN107179079A (en) * | 2017-05-29 | 2017-09-19 | 桂林电子科技大学 | The indoor orientation method merged based on PDR with earth magnetism |
| CN107504971A (en) * | 2017-07-05 | 2017-12-22 | 桂林电子科技大学 | A kind of indoor orientation method and system based on PDR and earth magnetism |
-
2018
- 2018-04-08 CN CN201810305986.1A patent/CN108592914A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927739A (en) * | 2014-01-10 | 2014-07-16 | 北京航天飞行控制中心 | Patroller positioning method based on spliced images |
| US20160360501A1 (en) * | 2014-12-18 | 2016-12-08 | Alibaba Group Holding Limited | Method and apparatus of positioning mobile terminal based on geomagnetism |
| CN105222772A (en) * | 2015-09-17 | 2016-01-06 | 泉州装备制造研究所 | A kind of high-precision motion track detection system based on Multi-source Information Fusion |
| CN105424030A (en) * | 2015-11-24 | 2016-03-23 | 东南大学 | Fusion navigation device and method based on wireless fingerprints and MEMS sensor |
| CN106767828A (en) * | 2016-12-29 | 2017-05-31 | 南京邮电大学 | A kind of mobile phone indoor positioning solution |
| CN106780699A (en) * | 2017-01-09 | 2017-05-31 | 东南大学 | A kind of vision SLAM methods aided in based on SINS/GPS and odometer |
| CN107179079A (en) * | 2017-05-29 | 2017-09-19 | 桂林电子科技大学 | The indoor orientation method merged based on PDR with earth magnetism |
| CN107144281A (en) * | 2017-06-30 | 2017-09-08 | 飞智控（天津）科技有限公司 | Unmanned plane indoor locating system and localization method based on cooperative target and monocular vision |
| CN107504971A (en) * | 2017-07-05 | 2017-12-22 | 桂林电子科技大学 | A kind of indoor orientation method and system based on PDR and earth magnetism |
Non-Patent Citations (14)
| Title |
|---|
| 任治新等: "基于改进ICP算法的地磁图匹配技术", 《计算机应用》 * |
| 刘传凯等: "嫦娥三号巡视器的惯导与视觉组合定姿定位", 《飞行器测控学报》 * |
| 吴志添等: "《面向水下地磁导航的地磁测量误差补偿方法研究》", 28 February 2017, 国防工业出版社 * |
| 张晓明等: "基于克里金插值的局部地磁图的构建", 《电子测量技术》 * |
| 张爱军等: "《导航定位技术及应用》", 30 June 2014, 电子科技大学出版社 * |
| 张聪聪等: "基于地磁场的室内定位和地图构建", 《仪器仪表学报》 * |
| 杨斌等: "基于标准化自协方差相关函数的EMD改进算法", 《计算机技术与发展》 * |
| 杨立溪: "《惯性技术手册》", 31 December 2013, 中国宇航出版社 * |
| 林雪原: "《组合导航及其信息融合方法》", 30 November 2017, 国防工业出版社 * |
| 王保丰等: "嫦娥三号巡视器视觉定位方法", 《中国科学: 信息科学》 * |
| 胡小平等: "《导航技术基础》", 31 July 2015, 国防工业出版社 * |
| 袁树友: "《上曜星月 中国北斗100问》", 31 May 2017, 解放军出版社 * |
| 赵琳等: "《船舶零航速减摇控制装置与系统》", 31 August 2015, 国防工业出版社 * |
| 陈慧岩等: "《航天器多源信息融合自主导航技术》", 31 March 2018, 北京理工大学出版社 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112147995A (en) * | 2019-06-28 | 2020-12-29 | 深圳市创客工场科技有限公司 | Robot motion control method and device, robot and storage medium |
| CN112147995B (en) * | 2019-06-28 | 2024-02-27 | 深圳市创客工场科技有限公司 | Robot motion control method and device, robot and storage medium |
| CN110648398A (en) * | 2019-08-07 | 2020-01-03 | 武汉九州位讯科技有限公司 | Real-time ortho image generation method and system based on unmanned aerial vehicle aerial data |
| CN110648398B (en) * | 2019-08-07 | 2020-09-11 | 武汉九州位讯科技有限公司 | Real-time ortho image generation method and system based on unmanned aerial vehicle aerial data |
| CN110455298A (en) * | 2019-08-14 | 2019-11-15 | 灵动科技（北京）有限公司 | Vehicle localization method and positioning system |
| CN113739784A (en) * | 2020-05-27 | 2021-12-03 | 华为技术有限公司 | Positioning method, user equipment, storage medium and electronic equipment |
| CN112013842A (en) * | 2020-08-29 | 2020-12-01 | 桂林电子科技大学 | Multi-mode indoor positioning method based on image geomagnetic field and inertial sensor |
| CN112560974A (en) * | 2020-12-22 | 2021-03-26 | 清华大学 | Information fusion and vehicle information acquisition method and device |
| CN113873442B (en) * | 2021-09-08 | 2023-08-04 | 宁波大榭招商国际码头有限公司 | Positioning method for external collection card |
| CN113873442A (en) * | 2021-09-08 | 2021-12-31 | 宁波大榭招商国际码头有限公司 | External hub card positioning method |
| CN114136309A (en) * | 2021-11-12 | 2022-03-04 | 上汽通用五菱汽车股份有限公司 | Positioning method, system and device of unmanned logistics vehicle and storage medium |
| CN114111808A (en) * | 2021-11-30 | 2022-03-01 | 上汽通用五菱汽车股份有限公司 | Positioning method, system and device of unmanned vehicle and readable storage medium |
| CN114117113A (en) * | 2022-01-28 | 2022-03-01 | 杭州宏景智驾科技有限公司 | Multi-feature-point motor vehicle positioning method and device, electronic equipment and storage medium |
| CN114117113B (en) * | 2022-01-28 | 2022-06-10 | 杭州宏景智驾科技有限公司 | Multi-feature-point motor vehicle positioning method and device, electronic equipment and storage medium |
| CN116698018A (en) * | 2023-08-08 | 2023-09-05 | 山西戴德测控技术股份有限公司 | Navigation positioning auxiliary device and coal mine tunnel navigation positioning system |
| CN116698018B (en) * | 2023-08-08 | 2023-10-13 | 山西戴德测控技术股份有限公司 | Navigation positioning auxiliary device and coal mine tunnel navigation positioning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108592914A (en) | The positioning of complex region inspecting robot, navigation and time service method under no GPS scenario | |
| CN104729506B (en) | A kind of unmanned plane Camera calibration method of visual information auxiliary | |
| Li et al. | High-fidelity sensor modeling and self-calibration in vision-aided inertial navigation | |
| Panahandeh et al. | Vision-aided inertial navigation based on ground plane feature detection | |
| Veth et al. | Fusion of low-cost imagining and inertial sensors for navigation | |
| CN110095116A (en) | A kind of localization method of vision positioning and inertial navigation combination based on LIFT | |
| US6489922B1 (en) | Passive/ranging/tracking processing method for collision avoidance guidance and control | |
| CN107796391A (en) | A kind of strapdown inertial navigation system/visual odometry Combinated navigation method | |
| CN109141433A (en) | A kind of robot indoor locating system and localization method | |
| CN105953795B (en) | A kind of navigation device and method for the tour of spacecraft surface | |
| CN103649680A (en) | Sensor positioning for 3D scanning | |
| CN107144278B (en) | Lander visual navigation method based on multi-source characteristics | |
| CN109669474B (en) | Priori knowledge-based multi-rotor unmanned aerial vehicle self-adaptive hovering position optimization algorithm | |
| KR101737950B1 (en) | Vision-based navigation solution estimation system and method in terrain referenced navigation | |
| Troni et al. | Magnetometer bias calibration based on relative angular position: Theory and experimental comparative evaluation | |
| Karam et al. | Integrating a low-cost mems imu into a laser-based slam for indoor mobile mapping | |
| Xian et al. | Fusing stereo camera and low-cost inertial measurement unit for autonomous navigation in a tightly-coupled approach | |
| Pan et al. | An optical flow-based integrated navigation system inspired by insect vision | |
| Briskin et al. | Estimating pose and motion using bundle adjustment and digital elevation model constraints | |
| Qian et al. | Optical flow based step length estimation for indoor pedestrian navigation on a smartphone | |
| Qian et al. | Optical flow-based gait modeling algorithm for pedestrian navigation using smartphone sensors | |
| Li et al. | Image-based self-position and orientation method for moving platform | |
| Huntsberger et al. | Sensory fusion for planetary surface robotic navigation, rendezvous, and manipulation operations | |
| Popov et al. | UAV navigation on the basis of video sequences registered by onboard camera | |
| Panahandeh et al. | IMU-camera data fusion: Horizontal plane observation with explicit outlier rejection |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |