CN111024091A - Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle - Google Patents
Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle Download PDFInfo
- Publication number
- CN111024091A CN111024091A CN201911415820.6A CN201911415820A CN111024091A CN 111024091 A CN111024091 A CN 111024091A CN 201911415820 A CN201911415820 A CN 201911415820A CN 111024091 A CN111024091 A CN 111024091A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- angle
- corridor
- attitude
- 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
Images
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
- G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a three-dimensional attitude algorithm for indoor flight of a vision-assisted micro unmanned aerial vehicle, which can realize pure autonomous calculation of a pitch angle, a roll angle and a course angle when the unmanned aerial vehicle flies indoors. When the magnetometer data is effective, the resolving of the three-dimensional attitude of the unmanned aerial vehicle is realized by utilizing the data of the IMU and the magnetometer, and the decoupling of the resolving of the horizontal attitude angle and the resolving of the course angle is realized; when magnetometer data fails, three-dimensional attitude calculation is realized only by IMU data, the influence of an accelerometer on course angle calculation is cut off, and the requirement of accuracy in endurance time is maintained when course angle calculation is carried out without the assistance of a magnetometer; when the unmanned aerial vehicle flies in an indoor corridor, the narrow-passage detection is realized through the front-view monocular camera module, the angle of the unmanned aerial vehicle deviating from a corridor line is extracted, and the heading angle accumulated error of the unmanned aerial vehicle is corrected through extended Kalman filtering by using the deviation angle.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicles, in particular to a three-dimensional attitude algorithm for indoor flight of a vision-assisted micro unmanned aerial vehicle.
Background
The micro unmanned aerial vehicle is very suitable for flight application of indoor environment, such as indoor detection, search and rescue, environment detection and the like, due to the characteristics of small size, light weight and concealment. However, the unmanned drone cannot adopt a navigation sensor with higher requirements on volume, weight and power consumption and a vision navigation algorithm with large calculation amount to realize indoor autonomous flight due to the limitation of the volume, weight and calculation capability of the drone. Meanwhile, the indoor magnetic field environment is complex, the magnetometer is greatly limited in indoor application, the attitude calculation of the unmanned aerial vehicle is interfered, the precision is reduced, and even the course angle calculation is wrong. Therefore, when the micro unmanned aerial vehicle flies indoors, the calculation of the three-dimensional attitude angle is also a key problem to be solved.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: in order to overcome the defects in the prior art, the invention provides a visual auxiliary three-dimensional attitude algorithm for indoor flight of a micro unmanned aerial vehicle.
The technical scheme adopted for solving the technical problems is as follows: a three-dimensional attitude algorithm for indoor flight of a vision-assisted micro unmanned aerial vehicle comprises resolving of a three-dimensional attitude angle of the unmanned aerial vehicle in a normal magnetic field environment, and decoupling of resolving of a horizontal attitude angle and a course angle in a resolving process; resolving a three-dimensional attitude angle of the unmanned aerial vehicle in the magnetic field abnormal environment; correcting the visual auxiliary course angle of corridor flight; the method specifically comprises the following steps:
s1: the head of the unmanned aerial vehicle is parallel to any indoor corridor direction before taking off, and is electrified and initialized; obtaining output data of the IMU and the three-axis magnetometer; meanwhile, the vision module starts to detect the front corridor by using a narrow corridor detection method and calculates the angle of the unmanned aerial vehicle deviating from the corridor line; wherein the vision module comprises a forward looking monocular camera and an image processing computer.
S2: calculating the geomagnetic induction intensity of the departure point according to the output of the magnetometer, comparing the geomagnetic induction intensity value with a set range, judging that the magnetic field of the flight environment is normal when the geomagnetic induction intensity of the departure point is within the set range, entering a flight state by taking the magnetic heading given by the magnetometer as an initial heading angle, and executing S3; when the geomagnetic induction intensity at the departure point is out of the set range, determining that the magnetic field of the flight environment is abnormal, setting the initial heading angle to be 0, entering a flight state, and executing S4;
s3: calculating the geomagnetic induction intensity in flight during the flight process of the unmanned aerial vehicle in the normal magnetic field environment, judging whether the geomagnetic induction intensity in the flight process is within a set range, if not, judging that the magnetic field of the flight environment is converted from normal to abnormal, and entering the step S4; if the attitude angle is within the set range, measuring the three-dimensional attitude angle of the unmanned aerial vehicle by using a 9-axis sensor which is the IMU and the magnetometer together, decoupling the horizontal attitude angle calculation and the course angle calculation, and then entering the step S5;
s4: resolving an attitude quaternion and a three-dimensional attitude angle only by utilizing IMU data, and then entering a step S5;
s5: reading the angle of the unmanned aerial vehicle, which is output by the vision module in the step S1, deviating from the corridor line, judging whether the vision course measurement is updated or not, and if the vision course measurement is updated, performing EKF data fusion to realize course angle correction and outputting a three-dimensional posture; and if the three-dimensional attitude angle is not updated, directly outputting the three-dimensional attitude angle.
Further, the process of calculating the angle of the unmanned aerial vehicle deviating from the corridor line in the step S1 specifically includes: when the unmanned aerial vehicle flies in the corridor, a forward view image is obtained by photographing the corridor in front through the forward view monocular camera, the forward view image is sent to the image processing computer, and the image processing computer realizes narrow-lane detection through Hough transformation by utilizing the forward view image, so that a linear equation of a corridor line in a pixel coordinate system is obtained; and calculating the angle of the unmanned aerial vehicle deviating from the corridor line through a triangular relation by combining the pitch angle data of the unmanned aerial vehicle and the internal parameters obtained by calibrating the camera.
Further, the step of determining whether the magnetic field is normal in step S2 includes the steps of:
step S2.1: calibrating the output of the magnetometer to obtain the normal geomagnetic induction intensity HEAnd is combined with HE+/-5% of the geomagnetic induction intensity is used as a set range of the geomagnetic induction intensity, and the set range is stored in the unmanned aerial vehicle;
step S2.2: after the unmanned aerial vehicle is powered on and operated, calculating the arithmetic square root of the sum of squares output by the three-axis magnetometer in operation to obtain the magnetic induction H measured by the three-axis magnetometer in the flying environment;
step S2.3: unmanned aerial vehicle uses triaxial magnetic forceMagnetic induction H measured by meter and setting range H of geomagnetic inductionEComparison was made at. + -. 5% when H is at HEMagnetometer data are available within + -5%; h is not at HEMagnetometer data are not available in the + -5% range.
Further, the calibration method of the magnetometer in step S2.1 is: in an open space with small indoor magnetic field interference, the magnetometer is calibrated by adopting a horizontal 360-degree rotation calibration method, after calibration is completed, the arithmetic square root of the sum of squares output by the calibrated three-axis magnetometer is calculated at a calibration place, namely the normal geomagnetic induction intensity HEAnd is combined with HEAnd +/-5% is used as a set range of geomagnetic induction intensity.
Further, in step S3, the three-dimensional attitude angle of the unmanned aerial vehicle is measured by using the 9-axis sensor with the IMU and the magnetometer, and the horizontal attitude angle solution and the course angle solution are decoupled, wherein the calculation process is as follows:
normalizing the output of the accelerometer in the IMU, and recording as [ a ]xkaykazk]TCalculating the current k time angular velocity correction amount eak＝[eakxeakyeakz]T：
Wherein, [ 001 ]]TRepresents the normalized acceleration of gravity,
is the attitude matrix of the navigation coordinate system to the carrier coordinate system.
Normalizing the magnetometer output, and recording as [ m ]xkmykmzk]TCalculating the current k time angular velocity correction amount emk＝[emkxemkyemkz]T：
Wherein，[mEmNmU]TIs a normalized geomagnetic vector in a local geographic coordinate system.
By means of eakAnd emkReconstructing elements to obtain k time angular velocity correction value ekThe angular velocity correction value can realize decoupling of calculation of horizontal attitude and heading angle:
by means of ekCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
wherein, [ g ]xgygz]TOriginal angular velocity data output by a gyroscope in the IMU; kiIs an integral coefficient; kpIs a scaling factor.
Solving quaternion differential equation by using corrected angular velocity of gyroscope and adopting Runge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
Further, in step S4, the three-dimensional attitude angle of the unmanned aerial vehicle is calculated by using only the data of the 3-axis gyroscope and the 3-axis accelerometer of the IMU, and the calculation process is as follows:
e in formula (1)akzAt 0, suppression of the increase of the heading angle error is achieved, that is:
by means of eakCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
solving quaternion differential equation by using the corrected angular velocity of the gyroscope and adopting a Longge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
By utilizing the angle of the unmanned aerial vehicle deviating from the corridor line and adopting EKF as a data fusion tool, the course angle correction calculation process of the unmanned aerial vehicle is realized as follows:
when unmanned aerial vehicle corridor flies, the narrow passage detection is realized through Hough transformation, thereby obtaining the linear equation of the corridor line under the pixel coordinate system with the center of the pixel plane as the origin:
v＝mu+n (7)
wherein u and v are coordinates of the u axis and the v axis of the pixel coordinate system, respectively, m is the slope of the straight line, and n is the intercept of the straight line on the v axis.
Because the unmanned aerial vehicle belongs to the low head flight before flying, make its optical axis and the preceding parallel of this body coordinate system of unmanned aerial vehicle when the forward looking camera installation. The origin O of the camera coordinate system is the optical center, ZcThe axis coinciding with the optical axis and pointing in front of the camera, XcAxis directed to right side of camera body, YcThe camera depression angle α is calculated by the pitch angle theta calculated by the unmanned aerial vehicle attitude and heading system, α is-theta, β is the angle of the unmanned aerial vehicle deviating from the corridor line, d is the distance between the unmanned aerial vehicle and the corridor line, for no loss of generality, no matter whether the optical axis of the camera deviates to the left side or the right side of the corridor, the P is set as any point on the corridor line, the coordinate of the P point in the camera coordinate system is (a ', b, c), obviously when the optical axis of the camera deviates to the left side corridor, the P point is the point on the left side corridor line, a ' is greater than 0, a is the absolute value of a ', and the triangle relation can be obtained:
a pixel coordinate system is established in the center of a pixel plane imaged by a camera, the origin is the center O' of the pixel plane, and the horizontal and vertical coordinate axes are u and v respectively. P' is the corresponding pixel point of P point in the image, and the pixel coordinate is (u, v), then there is according to camera aperture principle of imaging then:
u＝fxa/c，v＝fyb/c (10)
in the formula (f)xAnd fyAnd parameters related to the focal length in the parameter matrix in the forward-looking camera are obtained through camera calibration. The equation of a straight line in equation (7) can be expressed as:
fyb/c＝mfxa/c+n (11)
substitution of formulae (8) and (9) into formula (11), elimination of a and b, retention of only c, and work-up yields:
since the point P is an arbitrary point on the straight line of the throat, and the above equation holds for an arbitrary value of c, it can be obtained:
ncosαcosβ+fysinαcosβ-mfxsinβ＝0 (13)
the angle β at which the drone deviates from the corridor line can be calculated from the above equation:
the final output polarity of the angle β at which the drone deviates from the corridor line is defined as β, which is output as a positive value when the optical axis of the forward-looking camera is oriented toward the corridor line on the left side of the corridor, and β, which is output as a negative value when the optical axis of the forward-looking camera is oriented toward the corridor line on the right side of the corridor.
When the unmanned aerial vehicle flies in the corridor, the heading psi of the unmanned aerial vehicle is corrected by using the angle β deviating from the corridor line, which is obtained by visual calculation, the data fusion method adopts Kalman filtering, and the state equation is as follows:
in the formula (I), the compound is shown in the specification,
is the system state, ωx、ωy、ωzFor the output angular velocity of the gyroscope, q0～q3Is an attitude quaternion, wqIs the system process noise.
The measurement equation is as follows:
wherein z represents a measurement; v represents the measurement noise; psicThe heading angle, which is calculated for vision, is determined from the angle β at which the drone deviates from the corridor line and the corridor orientation.
The heading angle psi of the unmanned aerial vehicle is defined as north and west being positive and in the range of 0-2 pi. When the geomagnetic induction intensity of the starting point of the unmanned aerial vehicle is out of the set range, the corridor direction of the initial orientation of the unmanned aerial vehicle is defined as 0 degree, namely the corridor direction is determined as the pseudo north direction, the east-west and south-north directions used for attitude calculation in the flight process all refer to the pseudo east-west and south-north directions, and the visual heading measured value psi is measured under the conditioncIs defined as:
d) when the unmanned plane flies to the north along the north-south corridor, β is more than 0, psic＝2π-β；β＜0,ψc＝-β；
e) When the unmanned aerial vehicle flies south to south along the north-south corridor: psic＝π-β；
f) When the drone is flying east-west along the east-west corridor:
when the drone is flying westward along the east-west corridor:
when the geomagnetic induction intensity of the takeoff point of the unmanned aerial vehicle is within a set range, the coordinate adopted in the whole flight process is in the true north direction, and the unmanned aerial vehicle needs to be initially aligned to the magnetism of the corridorHeading, and the geometric relationship of the other corridors to the initially aligned corridor, converts the angle β off the corridor line to a visual heading measurement ψ of plus 0-2 π north off westc。
And taking the attitude quaternion obtained in the S2 and the S3 as one-step prediction of the attitude quaternion in the EKF course correction algorithm.
The kalman filter matrix is as follows:
H＝[Hq0Hq1Hq2Hq3](19)
wherein:
continuous system discretization:
Φ≈I+FT；Γ＝GT (24)
after the calculation, the EKF is used for completing data fusion to obtain the corrected attitude quaternion, thereby realizing the correction of the course angle.
The invention has the beneficial effects that: according to the vision-assisted three-dimensional attitude algorithm for the indoor flight of the micro unmanned aerial vehicle, normal calculation of three-dimensional attitude of the micro unmanned aerial vehicle in all magnetic field environments during indoor flight is realized, decoupling of horizontal attitude angle calculation and course angle calculation is realized, and calculation accuracy of course angles is improved; and correcting the error of the heading angle of the unmanned aerial vehicle by the result of the vision narrow-passage detection algorithm with small calculation amount. The method is small in calculation amount, high in precision and suitable for resolving the indoor flight attitude of the micro unmanned aerial vehicle.
Drawings
The invention is further illustrated by the following figures and examples.
FIG. 1 is a schematic diagram of a three-dimensional attitude solution process according to the present invention.
Fig. 2 is a schematic diagram of coordinate system definition in corridor flight unmanned aerial vehicle corridor line deviation angle extraction.
Fig. 3 is a schematic diagram of a pixel coordinate system definition with the center of a pixel plane as an origin.
Detailed Description
The present invention will now be described in detail with reference to the accompanying drawings. This figure is a simplified schematic diagram, and merely illustrates the basic structure of the present invention in a schematic manner, and therefore it shows only the constitution related to the present invention.
As shown in FIG. 1, the three-dimensional attitude algorithm for the indoor flight of the vision-assisted micro unmanned aerial vehicle comprises the steps of resolving a three-dimensional attitude angle of the unmanned aerial vehicle in a normal magnetic field environment, and decoupling the horizontal attitude angle and the course angle in the resolving process; resolving a three-dimensional attitude angle of the unmanned aerial vehicle in the magnetic field abnormal environment; correcting the visual auxiliary course angle of corridor flight; in this embodiment, the vision module is a front-view monocular camera, and the images collected by the camera are sent to an image processing computer for processing.
The three-dimensional attitude algorithm specifically comprises the following steps:
s1: the orientation of a nose of the unmanned aerial vehicle before takeoff is parallel to any indoor corridor direction, power-on initialization is carried out, and output data of an IMU and a three-axis magnetometer are obtained; meanwhile, the front-view monocular camera starts to take a picture of the front corridor, the picture is sent to the image processing computer, corridor detection is achieved by using a narrow corridor detection method, and the angle of the unmanned aerial vehicle deviating from a corridor line is calculated.
S2: calculating the geomagnetic induction intensity of the flying starting point according to the output of the magnetometer, comparing the geomagnetic induction intensity value with a set range, if the geomagnetic induction intensity is within the set range, determining that the magnetic field of the flying environment is normal, taking the magnetic heading angle given by the magnetometer as an initial heading angle, entering a flying state, and executing S3; if the geomagnetic induction intensity at the departure point is out of the set range, determining that the magnetic field of the flight environment is abnormal, setting the initial heading angle to be 0, entering a flight state, and executing S4; under the condition of abnormal environmental magnetic field, the attitude calculation is carried out in the whole flight process without using magnetometer data; when the unmanned aerial vehicle enters the corridor to fly, the forward-looking monocular camera performs the narrow lane detection and calculates the angle of the unmanned aerial vehicle deviating from the corridor line, the heading angle of the unmanned aerial vehicle is corrected by using the angle, and S5 is executed.
Wherein, judge whether magnetic field is normal, export calculation earth induction intensity through the magnetometer promptly, compare earth induction intensity with the settlement scope, the computational process is as follows:
step S2.1: before judgment, firstly, calibrating the output of the magnetometer to determine the size of a set range, wherein calibration is only needed once, the set range is stored in the unmanned aerial vehicle, and the data can be called when judgment is carried out; the specific calibration process comprises the following steps: in an open space with small indoor magnetic field interference, the magnetometer is calibrated by adopting a horizontal 360-degree rotation calibration method, after calibration is completed, the arithmetic square root of the sum of squares output by the calibrated three-axis magnetometer is calculated at a calibration place, namely the normal geomagnetic induction intensity HEAnd is combined with HEPlus or minus 5 percent is used as the set range of the geomagnetic induction intensity;
step S2.2: after the unmanned aerial vehicle is powered on and operated, calculating the arithmetic square root of the sum of squares output by the three-axis magnetometer in operation to obtain the magnetic induction H measured by the three-axis magnetometer in the flying environment;
step S2.3: unmanned aerial vehicle sets for scope H with magnetic induction intensity H and earth magnetism induction intensity of three axis magnetometer measurementEComparison was made at. + -. 5% when H is at HEMagnetometer data are available within + -5%; h is not at HEMagnetometer data are not available in the + -5% range.
S3: in the flight process of the unmanned aerial vehicle in the normal magnetic field environment, taking the magnetic heading as an initial heading angle, calculating the geomagnetic induction intensity in flight, judging whether the geomagnetic induction intensity in the flight process is in a set range, if not, judging that the magnetic field in the flight environment is converted from normal to abnormal, and entering the step S4; if the attitude angle is within the set range, measuring the three-dimensional attitude angle of the unmanned aerial vehicle by using a 9-axis sensor which is the IMU and the magnetometer together, decoupling the horizontal attitude angle calculation and the course angle calculation, and then entering the step S5;
wherein, the earth magnetism induction intensity that magnetometer measured in the flight is in setting for the within range, and the take-off point earth magnetism induction intensity is also in setting for the within range, will adopt 9 sensor data to calculate unmanned aerial vehicle's three-dimensional attitude angle to realize the decoupling zero that horizontal attitude was solved and the course angle is solved, the computational process is as follows:
normalizing the output of the accelerometer in the IMU, and recording as [ a ]xkaykazk]TCalculating the current k time angular velocity correction amount eak＝[eakxeakyeakz]T：
Wherein, [ 001 ]]TRepresents the normalized acceleration of gravity,
an attitude matrix from a navigation coordinate system to a carrier coordinate system;
normalizing the magnetometer output, and recording as [ m ]xkmykmzk]TCalculating the current k time angular velocity correction amount emk＝[emkxemkyemkz]T：
[mEmNmU]TGeomagnetism in a local geographical coordinate system as normalizedA vector;
by means of eakAnd emkReconstructing elements to obtain k time angular velocity correction value ekThe angular velocity correction value can realize decoupling of calculation of horizontal attitude and heading angle:
by means of ekCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
wherein, [ g ]xgygz]TOriginal angular velocity data output by a gyroscope in the IMU; kiIs an integral coefficient; kpIs a proportionality coefficient;
solving quaternion differential equation by using corrected angular velocity of gyroscope and adopting Runge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
S4: according to the judgment of the steps S2 and S3, when the geomagnetic induction intensity in flight is out of the set range, or the geomagnetic induction intensity in flight is in the set range, but the geomagnetic induction intensity at the flying point is out of the set range, the IMU data is only used for resolving the three-dimensional attitude angle, wherein the horizontal attitude angle is corrected by the accelerometer data, and the influence of the accelerometer on the heading angle is reduced to 0 by setting the compensation value of the yaw axis gyroscope to zero, so that the increase of the heading angle error is inhibited, and the heading accuracy requirement in the endurance time of the micro unmanned aerial vehicle is met; then proceeds to step S5;
wherein, when the geomagnetic induction intensity is outside the set range in flight, or the geomagnetic induction intensity is within the set range in flight, but the geomagnetic induction intensity at the point of departure is outside the set range, will only adopt 3-axis gyroscope and 3-axis accelerometer data of IMU to solve the three-dimensional attitude angle of unmanned aerial vehicle, the calculation process is as follows:
e in formula (1)akzAt 0, suppression of the increase of the heading angle error is achieved, that is:
by means of eakCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
solving quaternion differential equation by using the corrected angular velocity of the gyroscope and adopting a Longge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
S5: reading the angle of the unmanned aerial vehicle deviating from the corridor line calculated in the step S1, judging whether the visual course measurement is updated, namely judging whether the angle is changed, and if the angle is updated, performing EKF data fusion to realize course angle correction and outputting a three-dimensional posture; and if the three-dimensional attitude angle is not updated, directly outputting the three-dimensional attitude angle.
When the unmanned aerial vehicle flies in the corridor, the front corridor is photographed by a front-view monocular camera, and the image processing computer realizes the narrow corridor detection by using the front-view image through Hough transformation, so that a linear equation of a corridor line in a pixel coordinate system is obtained; and calculating the angle and distance information of the unmanned aerial vehicle deviating from the corridor line by utilizing the pitch angle data given by the unmanned aerial vehicle attitude and heading reference system, the linear equation of the corridor line in the pixel plane and the internal parameters obtained by calibrating the camera through the triangular relation. By utilizing the angle of the unmanned aerial vehicle deviating from the corridor line and adopting EKF as a data fusion tool, the course angle correction calculation process of the unmanned aerial vehicle is realized as follows:
when unmanned aerial vehicle corridor flies, the narrow passage detection is realized through Hough transformation, thereby obtaining the linear equation of the corridor line under the pixel coordinate system with the center of the pixel plane as the origin:
v＝mu+n (7)
wherein u and v are coordinates of the u axis and the v axis of the pixel coordinate system, respectively, m is the slope of the straight line, and n is the intercept of the straight line on the v axis.
Because the unmanned aerial vehicle belongs to the low head flight before flying, make its optical axis and the preceding parallel of this body coordinate system of unmanned aerial vehicle when the forward looking camera installation. As shown in FIG. 2, the origin O of the camera coordinate system is the optical center, ZcThe axis coinciding with the optical axis and pointing in front of the camera, XcAxis directed to right side of camera body, YcThe camera depression angle α is calculated by the pitch angle theta calculated by the unmanned aerial vehicle attitude and heading system, α is-theta, β is the angle of the unmanned aerial vehicle deviating from the corridor line, d is the distance between the unmanned aerial vehicle and the corridor line, for no loss of generality, no matter whether the optical axis of the camera deviates to the left side or the right side of the corridor, the P is set as any point on the corridor line, the coordinate of the P point in the camera coordinate system is (a ', b, c), obviously when the optical axis of the camera deviates to the left side corridor, the P point is the point on the left side corridor line, a ' is greater than 0, a is the absolute value of a ', and the triangle relation can be obtained:
as shown in fig. 3, a pixel coordinate system is established at the center of the pixel plane imaged by the camera, and the origin is the center O' of the pixel plane, and the horizontal and vertical axes are u and v, respectively. P' is the corresponding pixel point of P point in the image, and the pixel coordinate is (u, v), then there is according to camera aperture principle of imaging then:
u＝fxa/c，v＝fyb/c (10)
in the formula (f)xAnd fyAnd parameters related to the focal length in the parameter matrix in the forward-looking camera are obtained through camera calibration. The equation of a straight line in equation (7) can be expressed as:
fyb/c＝mfxa/c+n (11)
substitution of formulae (8) and (9) into formula (11), elimination of a and b, retention of only c, and work-up yields:
since the point P is an arbitrary point on the straight line of the throat, and the above equation holds for an arbitrary value of c, it can be obtained:
ncosαcosβ+fysinαcosβ-mfxsinβ＝0 (13)
the angle β at which the drone deviates from the corridor line can be calculated from the above equation:
the final output polarity of the angle β at which the drone deviates from the corridor line is defined as β, which is output as a positive value when the optical axis of the forward-looking camera is oriented toward the corridor line on the left side of the corridor, and β, which is output as a negative value when the optical axis of the forward-looking camera is oriented toward the corridor line on the right side of the corridor.
When the unmanned aerial vehicle flies in the corridor, the heading psi of the unmanned aerial vehicle is corrected by using the angle β deviating from the corridor line, which is obtained by visual calculation, the data fusion method adopts Kalman filtering, and the state equation is as follows:
in the formula (I), the compound is shown in the specification,
is the system state, ωx、ωy、ωzFor the output angular velocity of the gyroscope, q0～q3Is an attitude quaternion, wqIs the system process noise.
The measurement equation is as follows:
wherein z represents a measurement; v represents the measurement noise; psicFor the visionThe calculated heading angle is determined from the angle β at which the drone deviates from the corridor line and the corridor orientation.
The heading angle psi of the unmanned aerial vehicle is defined as north and west being positive and in the range of 0-2 pi. When the geomagnetic induction intensity of the starting point of the unmanned aerial vehicle is out of the set range, the corridor direction of the initial orientation of the unmanned aerial vehicle is defined as 0 degree, namely the corridor direction is determined as the pseudo north direction, the east-west and south-north directions used for attitude calculation in the flight process all refer to the pseudo east-west and south-north directions, and the visual heading measured value psi is measured under the conditioncIs defined as:
g) when the unmanned plane flies to the north along the north-south corridor, β is more than 0, psic＝2π-β；β＜0,ψc＝-β；
h) When the unmanned aerial vehicle flies south to south along the north-south corridor: psic＝π-β；
i) When the drone is flying east-west along the east-west corridor:
when the drone is flying westward along the east-west corridor:
when the geomagnetic induction intensity of the departure point of the unmanned aerial vehicle is within the set range, the coordinate adopted in the whole flight process is in the true north direction, and the magnetic heading of the corridor initially aligned by the unmanned aerial vehicle and the geometric relationship between other corridors and the initially aligned corridor need to be utilized to convert the angle β deviating from the corridor line into the visual heading measured value psi with the north and the west deviating from the positive 0-2 pic。
And taking the attitude quaternion obtained in the S2 and the S3 as one-step prediction of the attitude quaternion in the EKF course correction algorithm.
The kalman filter matrix is as follows:
H＝[Hq0Hq1Hq2Hq3](19)
wherein:
continuous system discretization:
Φ≈I+FT；Γ＝GT (24)
after the calculation, the EKF is used for completing data fusion to obtain the corrected attitude quaternion, thereby realizing the correction of the course angle.
The method can realize the fully autonomous calculation of the pitching angle, the roll angle and the course angle of the unmanned aerial vehicle under the normal and abnormal magnetic field environments, and the decoupling of the calculation of the horizontal attitude angle and the calculation of the course angle is realized. The algorithm adopts an MEMS IMU, a three-axis magnetometer and a forward-looking camera as data sources; the three-axis magnetometer outputs the magnitude of geomagnetic induction intensity to judge, when the magnitude of local magnetic induction intensity is within a set range, the algorithm utilizes the data of 9-axis sensors, namely the IMU and the magnetometer, to realize the resolving of the three-dimensional attitude of the unmanned aerial vehicle, and simultaneously realizes the decoupling of horizontal attitude angle resolving and course angle resolving, the accelerometer outputs only compensate the horizontal attitude angle, the magnetometer outputs only compensate the course angle, and the influence of magnetic field fluctuation on the horizontal attitude angle is avoided; when the local magnetic induction intensity is out of the set range, the algorithm does not use magnetometer data to correct the course angle, only uses accelerometer data to correct the horizontal attitude, reduces the influence of the accelerometer on the course angle to 0, inhibits the increase of the error of the course angle, and can solve the requirement of the accuracy of the unmanned aerial vehicle course angle in the duration time under the condition without the assistance of the magnetometer; when the unmanned aerial vehicle flies in an indoor corridor, the narrow lane detection is realized through the front-view monocular camera, the angle of the unmanned aerial vehicle deviating from a corridor line is extracted, the heading angle is corrected by using the deviation angle, and the problem of unmanned aerial vehicle heading error accumulation when the magnetic heading error is large or the magnetic heading is unavailable during indoor flying is solved. The algorithm realizes the full-automatic solution of the three-dimensional attitude angle of the unmanned aerial vehicle in indoor flight in all magnetic environments by means of a low-precision MEMS IMU (micro-electromechanical system), a three-axis magnetometer and a forward-looking monocular camera which are composed of a three-axis gyroscope and a three-axis accelerometer.
In light of the foregoing description of preferred embodiments in accordance with the invention, it is to be understood that numerous changes and modifications may be made by those skilled in the art without departing from the scope of the invention. The technical scope of the present invention is not limited to the contents of the specification, and must be determined according to the scope of the claims.
Claims (7)
1. The utility model provides a three-dimensional attitude algorithm of supplementary miniature unmanned aerial vehicle indoor flight of vision which characterized in that: the method comprises the steps of resolving the three-dimensional attitude angle of the unmanned aerial vehicle in the normal magnetic field environment, and decoupling the horizontal attitude angle and the course angle in the resolving process; resolving a three-dimensional attitude angle of the unmanned aerial vehicle in the magnetic field abnormal environment; correcting the visual auxiliary course angle of corridor flight; the method specifically comprises the following steps:
s1: the head of the unmanned aerial vehicle is parallel to any indoor corridor direction before taking off, and is electrified and initialized; obtaining output data of the IMU and the three-axis magnetometer; meanwhile, the vision module starts to detect the front corridor by using a narrow corridor detection method and calculates the angle of the unmanned aerial vehicle deviating from the corridor line;
s2: calculating the geomagnetic induction intensity of the departure point according to the output of the magnetometer, comparing the geomagnetic induction intensity value with a set range, judging that the magnetic field of the flight environment is normal when the geomagnetic induction intensity of the departure point is within the set range, entering a flight state by taking the magnetic heading given by the magnetometer as an initial heading angle, and executing S3; when the geomagnetic induction intensity at the departure point is out of the set range, determining that the magnetic field of the flight environment is abnormal, setting the initial heading angle to be 0, entering a flight state, and executing S4;
s3: calculating the geomagnetic induction intensity in flight during the flight process of the unmanned aerial vehicle in the normal magnetic field environment, judging whether the geomagnetic induction intensity in the flight process is within a set range, if not, judging that the magnetic field of the flight environment is converted from normal to abnormal, and entering the step S4; if the attitude angle is within the set range, measuring the three-dimensional attitude angle of the unmanned aerial vehicle by using a 9-axis sensor which is the IMU and the magnetometer together, decoupling the horizontal attitude angle calculation and the course angle calculation, and then entering the step S5;
s4: resolving an attitude quaternion and a three-dimensional attitude angle only by utilizing IMU data, and then entering a step S5;
s5: reading the angle of the unmanned aerial vehicle, which is output by the vision module in the step S1, deviating from the corridor line, judging whether the vision course measurement is updated or not, and if the vision course measurement is updated, performing EKF data fusion to realize course angle correction and outputting a three-dimensional posture; and if the three-dimensional attitude angle is not updated, directly outputting the three-dimensional attitude angle.
2. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicles according to claim 1, wherein: the process of calculating the angle of the unmanned aerial vehicle deviating from the corridor line in the step S1 specifically includes:
when the unmanned aerial vehicle flies in the corridor, a forward view image is obtained by photographing the corridor in front through the forward view monocular camera, the forward view image is sent to the image processing computer, and the image processing computer realizes narrow-lane detection through Hough transformation by utilizing the forward view image, so that a linear equation of a corridor line in a pixel coordinate system is obtained; and calculating the angle of the unmanned aerial vehicle deviating from the corridor line through a triangular relation by combining the pitch angle data of the unmanned aerial vehicle and the internal parameters obtained by calibrating the camera.
3. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicles according to claim 1, wherein: the determination of whether the magnetic field is normal in step S2 includes the steps of:
step S2.1: calibrating the output of the magnetometer to obtain the normal geomagnetic induction intensity HEAnd is combined with HE+/-5% of the geomagnetic induction intensity is used as a set range of the geomagnetic induction intensity, and the set range is stored in the unmanned aerial vehicle;
step S2.2: after the unmanned aerial vehicle is powered on and operated, calculating the arithmetic square root of the sum of squares output by the three-axis magnetometer in operation to obtain the magnetic induction H measured by the three-axis magnetometer in the flying environment;
step S2.3: unmanned aerial vehicle sets for scope H with magnetic induction intensity H and earth magnetism induction intensity of three axis magnetometer measurementEComparison was made at. + -. 5% when H is at HEMagnetometer data are available within + -5%; h is not at HEMagnetometer data are not available in the + -5% range.
4. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicles according to claim 3, wherein: the calibration method of the magnetometer in the step S2.1 comprises the following steps: in an open space with small indoor magnetic field interference, the magnetometer is calibrated by adopting a horizontal 360-degree rotation calibration method, after calibration is completed, the arithmetic square root of the sum of squares output by the calibrated three-axis magnetometer is calculated at a calibration place, namely the normal geomagnetic induction intensity HEAnd is combined with HEAnd +/-5% is used as a set range of geomagnetic induction intensity.
5. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicles according to claim 1, wherein: in the step S3, the three-dimensional attitude angle of the unmanned aerial vehicle is measured by using the 9-axis sensor with the IMU and the magnetometer, and the horizontal attitude angle solution and the course angle solution are decoupled, wherein the calculation process is as follows:
normalizing the output of the accelerometer in the IMU, and recording as [ a ]xkaykazk]TCalculating the current k time angular velocity correction amount eak＝[eakxeakyeakz]T：
Wherein, [ 001 ]]TRepresents the normalized acceleration of gravity,
an attitude matrix from a navigation coordinate system to a carrier coordinate system;
normalizing the magnetometer output, and recording as [ m ]xkmykmzk]TCalculating the current k time angular velocity correction amount emk＝[emkxemkyemkz]T：
Wherein [ m ]EmNmU]TIs a normalized geomagnetic vector under a local geographic coordinate system;
by means of eakAnd emkReconstructing elements to obtain k time angular velocity correction value ekThe angular velocity correction value can realize decoupling of calculation of horizontal attitude and heading angle:
by means of ekCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
wherein, [ g ]xgygz]TOriginal angular velocity data output by a gyroscope in the IMU; kiIs an integral coefficient; kpIs a proportionality coefficient;
solving quaternion differential equation by using corrected angular velocity of gyroscope and adopting Runge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
6. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicles according to claim 5, wherein: in the step S4, the three-dimensional attitude angle of the unmanned aerial vehicle is calculated by using only the data of the 3-axis gyroscope and the 3-axis accelerometer of the IMU, and the calculation process is as follows:
e in formula (1)akzAt 0, suppression of the increase of the heading angle error is achieved, that is:
by means of eakCorrecting the angular speed of the gyroscope to obtain the corrected angular speed output [ g ] of the gyroscope at the moment kxkgykgzk]TAnd (3) calculating:
solving quaternion differential equation by using the corrected angular velocity of the gyroscope and adopting a Longge Kutta method to realize attitude quaternion q0～q3And (4) updating to solve the three-dimensional attitude of the unmanned aerial vehicle.
7. The vision-aided three-dimensional attitude algorithm for indoor flight of micro unmanned aerial vehicle as claimed in any one of claims 1 to 6, wherein: by utilizing the angle of the unmanned aerial vehicle deviating from the corridor line and adopting EKF as a data fusion tool, the course angle correction calculation process of the unmanned aerial vehicle is realized as follows:
when unmanned aerial vehicle corridor flies, the narrow passage detection is realized through Hough transformation, thereby obtaining the linear equation of the corridor line under the pixel coordinate system with the center of the pixel plane as the origin:
v＝mu+n (7)
wherein u and v are the coordinates of the u axis and the v axis of the pixel coordinate system respectively, m is the slope of the straight line, and n is the intercept of the straight line on the v axis;
because the unmanned aerial vehicle belongs to low-head flight when flying forward, the optical axis of the forward-looking camera is parallel to the forward direction of the coordinate system of the unmanned aerial vehicle body when the forward-looking camera is installed; the origin O of the camera coordinate system is the optical center, ZcThe axis coinciding with the optical axis and pointing in front of the camera, XcAxis directed to right side of camera body, YcThe camera depression angle α is calculated by a pitch angle theta calculated by an unmanned aerial vehicle attitude and heading system, α is equal to-theta, P is any point on a corridor line, the coordinate of the point P in a camera coordinate system is (a', b, c), and the three coordinate axes are in a triangular relation:
wherein β is the angle of the unmanned aerial vehicle deviating from the corridor line, d is the distance between the unmanned aerial vehicle and the corridor line, and a is the absolute value of a';
establishing a pixel coordinate system in the center of a pixel plane imaged by a camera, wherein the origin is the center O' of the pixel plane, and the horizontal and vertical coordinate axes are u and v respectively; p' is the corresponding pixel point of P point in the image, and the pixel coordinate is (u, v), then there is according to camera aperture principle of imaging then:
u＝fxa/c，v＝fyb/c (10)
in the formula (f)xAnd fyParameters related to the focal length in the parameter matrix in the forward-looking camera are obtained through camera calibration;
the equation of a straight line in equation (7) can be expressed as:
fyb/c＝mfxa/c+n (11)
substitution of formulae (8) and (9) into formula (11), elimination of a and b, retention of only c, and work-up yields:
since the point P is an arbitrary point on the straight line of the throat, and the above equation holds for an arbitrary value of c, it can be obtained:
ncosαcosβ+fysinαcosβ-mfxsinβ＝0 (13)
the angle β at which the drone deviates from the corridor line can be calculated from equation (13) above:
the final output polarity of the angle β of the unmanned aerial vehicle deviating from the corridor line is defined as that β outputs a positive value when the optical axis of the front-view camera faces the corridor line on the left side of the corridor, and β converts the optical axis of the front-view camera into a negative value output when the optical axis of the front-view camera faces the corridor line on the right side of the corridor;
when the unmanned aerial vehicle flies in the corridor, the heading psi of the unmanned aerial vehicle is corrected by using the angle β deviating from the corridor line, which is obtained by visual calculation, the data fusion method adopts Kalman filtering, and the state equation is as follows:
in the formula (I), the compound is shown in the specification,
is the system state, ωx、ωy、ωzFor the output angular velocity of the gyroscope, q0～q3Is an attitude quaternion, wqIs the system process noise;
the measurement equation is as follows:
wherein z represents a measurement; v represents the measurement noise; psicDetermining the course angle obtained by vision calculation according to the angle β of the unmanned aerial vehicle deviating from the corridor line and the corridor orientation;
the heading angle psi of the unmanned aerial vehicle is defined as that north and west are positive and the range of 0-2 pi; when nobodyWhen the geomagnetic induction intensity of the takeoff point of the unmanned aerial vehicle is outside the set range, the corridor direction of the initial orientation of the unmanned aerial vehicle is defined as 0 degree, namely the corridor direction is determined as the pseudo north direction, the east-west and south-north directions used for attitude calculation in the flight process all refer to the pseudo east-west and south-north directions, and the visual heading measured value psi is measured under the conditioncIs defined as:
a) when the unmanned plane flies to the north along the north-south corridor, β is more than 0, psic＝2π-β；β＜0,ψc＝-β；
b) When the unmanned aerial vehicle flies south to south along the north-south corridor: psic＝π-β；
c) When the drone is flying east-west along the east-west corridor:
when the drone is flying westward along the east-west corridor:
when the geomagnetic induction intensity of the departure point of the unmanned aerial vehicle is within the set range, the coordinate adopted in the whole flight process is in the true north direction, and the angle β deviating from the corridor line is converted into the visual heading measured value psi with the north and the west being positive 0-2 pi by utilizing the magnetic heading of the unmanned aerial vehicle initially aligned to the corridor and the geometric relationship between other corridors and the initially aligned corridorc；
Taking the attitude quaternion obtained in S3 and S4 as one-step prediction of the attitude quaternion in the EKF course correction algorithm;
the kalman filter matrix is as follows:
H＝[Hq0Hq1Hq2Hq3](19)
wherein:
continuous system discretization:
Φ≈I+FT；Γ＝GT (24)
after the calculation, the EKF is used for completing data fusion to obtain the corrected attitude quaternion, thereby realizing the correction of the course angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911415820.6A CN111024091A (en) | 2019-12-31 | 2019-12-31 | Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911415820.6A CN111024091A (en) | 2019-12-31 | 2019-12-31 | Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111024091A true CN111024091A (en) | 2020-04-17 |
Family
ID=70197894
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911415820.6A Pending CN111024091A (en) | 2019-12-31 | 2019-12-31 | Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111024091A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111551175A (en) * | 2020-05-27 | 2020-08-18 | 北京计算机技术及应用研究所 | Complementary filtering attitude calculation method of attitude heading reference system |
| CN112346454A (en) * | 2020-10-28 | 2021-02-09 | 博康智能信息技术有限公司 | Unmanned ship control method and system based on neural network |
| CN112595317A (en) * | 2020-10-26 | 2021-04-02 | 一飞(海南)科技有限公司 | Unmanned aerial vehicle takeoff control method, system, medium, computer equipment and unmanned aerial vehicle |
| CN113740890A (en) * | 2021-08-31 | 2021-12-03 | 普宙科技(深圳)有限公司 | Course angle correction method, system, computer equipment and storage medium |
| CN118129695A (en) * | 2024-05-07 | 2024-06-04 | 浙江智强东海发展研究院有限公司 | Control method and chip of underwater acquisition device |
-
2019
- 2019-12-31 CN CN201911415820.6A patent/CN111024091A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111551175A (en) * | 2020-05-27 | 2020-08-18 | 北京计算机技术及应用研究所 | Complementary filtering attitude calculation method of attitude heading reference system |
| CN111551175B (en) * | 2020-05-27 | 2024-03-15 | 北京计算机技术及应用研究所 | Complementary filtering attitude resolving method of navigation attitude reference system |
| CN112595317A (en) * | 2020-10-26 | 2021-04-02 | 一飞(海南)科技有限公司 | Unmanned aerial vehicle takeoff control method, system, medium, computer equipment and unmanned aerial vehicle |
| CN112595317B (en) * | 2020-10-26 | 2024-05-24 | 一飞(海南)科技有限公司 | Unmanned aerial vehicle take-off control method, system, medium, computer equipment and unmanned aerial vehicle |
| CN112346454A (en) * | 2020-10-28 | 2021-02-09 | 博康智能信息技术有限公司 | Unmanned ship control method and system based on neural network |
| CN112346454B (en) * | 2020-10-28 | 2023-08-18 | 博康智能信息技术有限公司 | Unmanned ship control method and system based on neural network |
| CN113740890A (en) * | 2021-08-31 | 2021-12-03 | 普宙科技(深圳)有限公司 | Course angle correction method, system, computer equipment and storage medium |
| CN118129695A (en) * | 2024-05-07 | 2024-06-04 | 浙江智强东海发展研究院有限公司 | Control method and chip of underwater acquisition device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111024091A (en) | Three-dimensional attitude algorithm for indoor flight of vision-assisted micro unmanned aerial vehicle | |
| CN109931926B (en) | Unmanned aerial vehicle seamless autonomous navigation method based on station-core coordinate system | |
| CN106708066B (en) | View-based access control model/inertial navigation unmanned plane independent landing method | |
| CN108036785A (en) | A kind of aircraft position and orientation estimation method based on direct method and inertial navigation fusion | |
| CN112630813B (en) | Unmanned aerial vehicle attitude measurement method based on strapdown inertial navigation and Beidou satellite navigation system | |
| CN111351482B (en) | Multi-rotor aircraft combined navigation method based on error state Kalman filtering | |
| CN103697889B (en) | A kind of unmanned plane independent navigation and localization method based on multi-model Distributed filtering | |
| CN107806874B (en) | A kind of inertial navigation polar region Initial Alignment Method of vision auxiliary | |
| CN115574816B (en) | Bionic vision multi-source information intelligent perception unmanned platform | |
| CN106352897B (en) | It is a kind of based on the silicon MEMS gyro estimation error of monocular vision sensor and bearing calibration | |
| CN110736457A (en) | combination navigation method based on Beidou, GPS and SINS | |
| CN103712598A (en) | Attitude determination system and method of small unmanned aerial vehicle | |
| CN114910080B (en) | Three-dimensional navigation attitude determination method based on underwater downlink radiation light intensity and polarized light field | |
| CN112444245B (en) | Insect-imitating vision integrated navigation method based on polarized light, optical flow vector and binocular vision sensor | |
| CN111504323A (en) | Unmanned aerial vehicle autonomous positioning method based on heterogeneous image matching and inertial navigation fusion | |
| CN110793515A (en) | Unmanned aerial vehicle attitude estimation method based on single-antenna GPS and IMU under large-mobility condition | |
| CN110887478A (en) | Autonomous navigation positioning method based on polarization/astronomical assistance | |
| CN108444468B (en) | Directional compass integrating downward vision and inertial navigation information | |
| CN109407696B (en) | Course angle dynamic calibration method for unmanned aerial vehicle | |
| Li et al. | Combined RGBD-inertial based state estimation for MAV in GPS-denied indoor environments | |
| CN116182855B (en) | Combined navigation method of compound eye-simulated polarized vision unmanned aerial vehicle under weak light and strong environment | |
| CN114509071B (en) | Attitude measurement method for wind tunnel test model | |
| CN113834481B (en) | Night polarization angle error correction method based on starlight vector information | |
| CN116242343A (en) | Onboard high-precision optical fiber inertial measurement device and integrated navigation method | |
| CN116124079A (en) | Method and device for calculating true azimuth angle of carrier by utilizing sun |
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 |