• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The present invention relates to a method for calibrating magnetometers (20) equipping an object (1) evolving in an ambient magnetic field, the method being characterized in that it comprises steps of: (a) Acquisition - by magnetometers (20 ), of at least three components measured of the magnetic field at the level of the magnetometers (20), and - by inertial measurement means (11) integral with said object (1), with an angular speed of the object (1) ; (c) Determination by data processing means (21) of values of at least one calibration parameter of the magnetometers (20) minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation over the angular velocity of the object (1), - the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers (20), and - the at least one magnetic equation assuming a uniform and stationary character of the magnetic field at the magnetic measuring means (20).
    公开号:FR3082611A1
    申请号:FR1855160
    申请日:2018-06-13
    公开日:2019-12-20
    发明作者:David VISSIERE；Mathieu Hillion；Hendrik Meier
    申请人:Sysnav SAS；
    IPC主号:
    专利说明:

    Method for calibrating magnetometers fitted to an object
    GENERAL TECHNICAL AREA
    The present invention relates to the field of navigation without GNSS.
    More specifically, it relates to a method for calibrating magnetometers integral with a gyrometer.
    STATE OF THE ART
    It is now common to follow the position of a vehicle by GNSS (Global Navigation Satellite System, for example GPS) or by using a communication network (triangulation using radio stations, wifi network or others).
    These methods are very limited because they do not ensure availability and accuracy of information, both of which are affected by possible masking between the sources and the receiver. They also appear to be dependent on external technologies such as GNSS satellites which may be unavailable or even intentionally interfered with.
    Alternatively, there are also known "autonomous" methods for monitoring the relative movement of a vehicle in any environment using an inertial or magneto-inertial unit. By relative displacement, one understands the trajectory of the vehicle in space compared to a point and a reference given to initialization. In addition to the trajectory, these methods also make it possible to obtain the orientation of the vehicle relative to the same initial reference.
    A navigation class inertial unit is made up of at least three accelerometers and three gyrometers arranged in triaxis. Typically, gyrometers "maintain" a benchmark, in which double time integration of accelerometer measurements can estimate motion.
    It is notably known that in order to be able to use the conventional inertial navigation methods, such as implemented in heavy applications such as the navigation of fighter or airliners, submarines, ships, etc., it is necessary to use very high precision sensors.
    However, such very high precision sensors are expensive, heavy and bulky, and therefore unsuitable for consumer applications.
    Consequently, low-cost sensors are used, presenting significant risks of error, for which the calibration (process which makes it possible to identify and then remove faults in the measurement of the sensors) is therefore a critical process.
    Methods for calibrating magnetometers using a gyrometer are known.
    The request LIS2012 / 0116716 proposes to examine whether the evolution of a magnetic field between two measurement points corresponds to the change in attitude predicted by the integration of gyrometers. However, it makes the assumption that the magnetometric / accelerometric measurements are reliable, which is difficult when the system is embedded in a vehicle. Indeed, we observe in practice that:
    - the ambient magnetic field is not always homogeneous and stationary;
    - metallic elements (for example a bodywork) or nearby magnets have an impact on the magnetic measurement (soft iron and hard iron effects), see also application US2004 / 0123474.
    - To correctly calibrate a magnetometer, it is necessary to put it in a maximum of spatial orientation so as to excite the error model to have observability. This problem is easily solvable for equipment which can be carried by hand in the case where a calibration is carried out before using it (it is turned in all directions, see for example the application FR1757082). If the magnetometer equips a vehicle, we are forced:
    o by the fact that during a typical journey of a vehicle, roll and pitch only slightly deviate from zero.
    o by the fact that the body is magnetized, or the fact that the vehicle has been loaded, which means that the calibration may no longer be relevant; which will require regular recalibrations.
    Application LIS2011 / 0178707 relates more precisely to smartphone type devices comprising a magnetic compass and a gyrometer, and evokes the problem of magnetic disturbances. This request rather aims at calibrating the magnetic compass from gyrometric data, but also proposes the opposite, taking as for it a starting point a magneto-accelerometric quaternion q and obtaining the angular speed of registration from this quaternion <o = 2q “ 1 q. However, it is always assumed that the magnetometric / accelerometric measurements are reliable, and in the contrary case it is just proposed either to not update the bias of the gyrometer while this situation lasts, or to use a quaternion by default stored in a memory.
    It would be desirable to have a new method of calibrating the magnetometers of an object for the purpose of estimating the movement of this object which allows an excellent quality of result and is not restrictive.
    PRESENTATION OF THE INVENTION
    The present invention thus relates according to a first aspect to a method for calibrating magnetometers equipping an object evolving in an ambient magnetic field, the method being characterized in that it comprises steps of:
    (a) Acquisition
    - by magnetometers, at least three components measured of the magnetic field at the level of magnetometers, and
    - By inertial measurement means integral with said object, with an angular speed of the object;
    (c) Determination by data processing means of values of at least one magnetometer calibration parameter minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation on the angular speed of the object,
    the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers, and
    - at least one magnetic equation assuming a uniform and stationary character of the magnetic field at the level of the magnetic measurement means.
    According to other advantageous and non-limiting characteristics:
    • the estimated components of the magnetic field are linked to the measured components M (measurement) by a model M (measurement) = AM (.estimation ') + b magnet0 , where A and b magnet0 are the calibration parameters of the magnetometers;
    The method further a step (b) of estimating a parameter representative of an error on the calibration parameters, step (c) being implemented if said parameter representative of an error is greater than one predetermined threshold;
    • The method further comprises a step (d) of new estimation of said parameter representative of an error on the calibration parameters so as to distinguish an external magnetic disturbance from a change in magnetic properties of the object;
    The method further comprises, if at the end of step (d) said parameter representative of an error is less than a predetermined threshold, a step (e) of determining a subset of the attitudes of the 'object for which the calibration is relevant;
    • I step (c) includes the implementation of a recursive filter or an optimization;
    The inertial measurement means are a gyrometer, the angular speed of the object acquired in step (a) being a measured angular speed, and that used by the magnetic equation or equations is an angular speed estimated as a function of the angular speed measured and gyrometer calibration parameters, step (c) also comprising determining values of at least one gyrometer calibration parameter;
    • the estimated angular speed ω ^ ' ηΩίίοη) of the object is linked to the measured angular speed wj ™ e o sure) by a model œÿr ^ mation} = D + b gyro ), where D and b gyro are the parameters of gyrometer calibration;
    • the magnetic equation is of the form Μ = —ω x M, where M is the vector of the components of the magnetic field, and ω the angular velocity;
    • said expression is a function of | jif (estimate) _ | _ ω (.estimation) χ ^ (estimate) | .
    • said parameter representative of an error on the calibration parameters is either the average of | m (estimate) _ | _ ^ estimate) χ <2 M (estimate) | suf a j nterva | e of given time is a grad j ent spatja | said components of the magnetic field;
    • the method comprises a step (f) of estimation by the data processing means of the movement of said object as a function of the angular speed of the object, of the measured components of the magnetic field, and of the values of the calibration parameters.
    According to a second aspect, an object is proposed evolving in an ambient magnetic field, comprising inertial measurement means configured to acquire an angular speed of the object, magnetometers configured to acquire at least three components of the magnetic field, the object being characterized in that it further comprises data processing means configured for:
    - determine values of at least one magnetometer calibration parameter minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation 30 on the angular speed of the object,
    the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers, and
    - at least one magnetic equation assuming a uniform and stationary character of the magnetic field at the level of the magnetic measurement means.
    According to a third and a fourth aspect, a computer program product is proposed comprising code instructions for the execution of a method according to the first aspect of calibration of magnetometers; and storage means readable by computer equipment on which a computer program product comprises code instructions for the execution of a method according to the first aspect of magnetometer calibration.
    PRESENTATION OF THE FIGURES
    Other characteristics and advantages of the present invention will appear on reading the following description of a preferred embodiment. This description will be given with reference to the appended drawings in which:
    - Figure 1 shows an example of vehicle architecture for the implementation of the method according to the invention;
    - Figure 2 is a diagram showing the steps of a preferred embodiment of the method according to the invention.
    DETAILED DESCRIPTION
    Architecture
    With reference to FIG. 1, the present method allows the calibration of magnetometers 20 of an object 1 evolving in an ambient magnetic field (typically the earth's magnetic field, if necessary altered by nearby metallic objects), noted M. Comme already explained, the magnetic field is a vector field in this case three-dimensional, that is to say associating a vector of dimension three with each three-dimensional point in which the object 1 is mobile.
    This object 1 can be any mobile object whose knowledge of the position is desired, for example a vehicle, in particular a wheeled vehicle, a drone, etc., but also a person or a part of the body of this person (his hands, his head, etc.).
    The magnetometer (s) 20 are integral with the object 1, and are "axial", that is to say capable of measuring a component of said magnetic field, i.e. the projection of said magnetic field vector M along their axis.
    The magnetometers 20 are at least three in number so as to be able to acquire 3 components of the magnetic field. Advantageously, the magnetometers 20 are even at least 8 or even 9, advantageously organized in groups of three into "thaxes", ie a triplet of magnetometers 20 two by two orthogonal associated with the same spatial position and measuring the magnetic field according to the three axes.
    Preferably, the orthonormal coordinate system associated with the object is chosen by convention (and for ease in the rest of this description) such that the thaxes are advantageously oriented in accordance with said orthonormal coordinate system, so as to further facilitate the calculations.
    But the skilled person will in any case know how to transpose to any spatial arrangement of magnetometers.
    The object 1 is also equipped with inertial measurement means 11 capable of measuring the angular speed of the object (for example integral with the bodywork in the case of a vehicle, and generally fixed in the frame of reference of the object 1) according to a system of three orthogonal axes, which define the object frame. The means 11 preferably consist of one or more gyros, but can also constitute one or more gyroscopes or any other source of attitude or angular speed (3D).
    In the preferred embodiment where the object 1 is a vehicle, the rotation around the vertical axis of the vehicle is described by the angle on which the driver acts by turning the steering wheel. On generally flat ground, the changes in direction of the vehicle are in the horizontal plane, i.e. also along said vertical axis. In reality, non-zero values for the roll (rotation along the longitudinal axis of the vehicle) and the pitch (rotation along the transverse axis of the vehicle) can be the result for example of a sloping road but are typically low. .
    The object 1 can also advantageously be equipped with “additional” acquisition means 10 of a measured linear speed of the object 1 (denoted V), that is to say of the displacement. These means 10 can directly or indirectly make it possible to obtain the linear speed, and thus be of many types, for example inertial measurement means. Thus the gyrometer (s) 11 may be supplemented by one or more accelerometers, or even the object 1 may comprise an inertial unit with at least three accelerometers and three gyrometers arranged in thax.
    Alternatively, the means 10 may consist, if the object 1 is a wheeled vehicle, of at least two odometers each for one wheel of the vehicle, for example the two rear wheels, as shown in the example in FIG. 1 We note that a set of odometers is a simple and reliable means of obtaining a simple linear speed.
    The object 1 also comprises, as explained, processing means 21 (typically a processor) for the direct implementation in real time of the treatments of the present method, for example a vehicle on-board computer, and possibly a memory 22 , and an interface 23 for restoring information of the movement of the object 1 (an instantaneous speed value, a heading, a position on a map, etc.), and / or sending commands to the object 1. The object 1 can for this reason be in particular an autonomous vehicle, and the processing means 21 configured to implement the autonomous navigation of the vehicle. Thus, said commands are sent to the vehicle control bodies (engine, steering wheel actuators, etc.) so as to simulate driving by the driver.
    Note that the processing means 21 can be external to the object 1, and for example connected to the latter by a wireless network. Alternatively, the inertial measurement means 11 and the magnetometers 20 can be connected to the data processing means 21 in particular by wire, for example via Ethernet.
    Process
    The present method is a method for calibrating at least the magnetometers 20. By calibration is meant the determination of one or more calibration parameters, a list of which will be seen below. In particular, some calibration parameters can be considered reliable and predetermined, and others to be determined. As far as those to be determined are concerned, we can foresee that they present "current" values (in other words that a calibration has already taken place), and that these values will be modified if necessary (in the event of new calibration).
    It is assumed that the magnetometers 20 in themselves are a priori perfectly calibrated (which means that all the magnetic disturbances linked to the magnetometers themselves are identified and corrected), this is only their direct environment (bodywork, etc. .) which must be corrected in calibration in order to be able to estimate the earth's magnetic field that would be measured in the absence of object 1.
    In a particularly preferred embodiment, the method can also be a method for calibrating the inertial measurement means 11, i.e. the gyrometer (s) 11 and the magnetometers 20 can be calibrated simultaneously. This is an extremely advantageous mode, since as we will see there is no longer even need to assume that one of the gyrometer or magnetometer is reliable to use it as a reference for calibrating the other: the two automatically calibrate each other. Alternatively, it will of course be possible to consider the gyrometer 11 as properly calibrated and to calibrate the magnetometers 20 accordingly, which makes it possible for example to calibrate more parameters of the magnetometers 20.
    As will be seen, in an advantageous embodiment the present method is even a method of estimating the movement of the object 1, ie it comprises, after calibration, the use of the calibrated measurements to deduce therefrom reliably one or more components of the movement.
    In a first step (a), the method comprises the acquisition by the inertial measurement means 11 of an angular speed of the object 1, denoted wj ™ r e o sure) , and by the magnetometers 20, of minus three components of the magnetic field. These components are more precisely so-called measured components, forming a vector denoted M (measurement) . Preferably, three components of the angular velocity are acquired and at least eight magnetic components (advantageously three magnetometric thaxes positioned at the corners of an isosceles rectangular triangle), so as to be able to deduce therefrom the field and its gradient (ie the derivatives at the following order along the three axes).
    These quantities are advantageously measured with a sampling dt (ie every "dt" seconds) with dt very small compared to the characteristic time of the movements of object 1, typically 40 ms.
    In a step (c) (as will be seen below the method advantageously comprises an intermediate step (b), but the latter remains optional), the data processing means 21 determine the values of at least one calibration parameter magnetometers 20 minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation on the angular velocity of the object 1.
    The idea is to separately estimate gyrometric data and magnetic data which are theoretically linked by a magnetic equation. Thus thanks to the equation we can express a quantity which ideally should be zero (i.e. the gyrometric and magnetic data exactly verify the magnetic equation), otherwise it is that the calibration is to be perfected.
    The estimated components of the magnetic field, denoted M (estimate) , are a function of the measured components of the magnetic field M (measurement) and of calibration parameters of the magnetometers 20.
    Preferably, they are linked to the measured components M (measurement) by a model M (measurement) = AM ^ estimate >> + b magnet0 , where the matrix 3x3 A and the vector b magnet0 are the calibration parameters of the magnetometers 20. Insofar as it is rather the estimated component that one wishes expressed as a function of the measured component, one can write = A -1
    - b magnet0 ).
    bmagneto is generally representative of “hard iron type” effects, and matrix A representative of “soft iron type” effects. These effects correspond to the hard / soft iron effects proper and can also include phenomena with the same impact but a different cause (eg effects due to electronics / physics of the magnetometer).
    Note that more generally, we can look at a general error model: M (estlmatl0n) = hfJAÎ ™ · 6511 ™ ')} with h a f onc tj on (application) which is not necessarily refined.
    In the linear case (M (measure) = AM ^ estimate >> + b magnet0 ), A is a 3x3 matrix, which can be written as A = R-Â with orthogonal R and upper triangular Â.
    Recall that it is assumed that the magnetometers 20 in themselves are a priori perfectly calibrated, it is only their direct environment (bodywork, etc.) which must be corrected in calibration in order to be able to estimate the terrestrial magnetic field that the 'we would measure in the absence of the object 1. The matrix R is thus important and to be determined so that the magnetic reference object remains stable.
    The classical methods of calibration with respect to spheres only determine R at a rotation around a near axis (see the document E. Dorveaux, D. Vissière, AP Martin, N. Petit, Iterative calibration method for inertial and magnetic sensors, in Proc, of the 48th IEEE Conf on Decision and Control 2009). The sensor axes do not remain fixed after the application of these methods.
    We can in practice determine 8 coefficients of A in calibration (a coefficient, the global scale factor which links the measurement to the physical unit in
    Tesla (or Gauss) cannot be determined, but, at the same time, it is not necessary to know it for a certain number of applications (ex: use of magnetometers to determine the course of an object). Note that it remains advantageously desirable to store the (average) value of the field standard during the calibration (relative to the selected scale factor), see below.
    As explained, the angular speed acquired in step (a) is preferably a measured speed (ie potentially marred by errors), and that used by said magnetic equation is an estimated speed, denoted M g e y ^ matlon as a function of the measured speed and of calibration parameters of the inertial measurement means 11.
    Preferably, in particular in the case where the means 11 are a gyrometer, in general ω ^ „ ηιαί: ιοη '> = g (<^ g ^ ure) ) with g a function (application) which is determined by the parameters calibration. In the simplest case, g is affine and the estimated angular speed Mgy s ^ l o matlon) is related to the angular speed measured wj ™ r e o sure) by the formula ω ^ ί ο Ι7ηαίΙοη) = D (Mg ™ ro Ure ^ + bgyro), where D and b gyro are the gyrometer 11 calibration parameters.
    In the case of a three-dimensional attitude, b gyro is a vector of bias and D is a 3x3 matrix (orthogonal matrix of passage to the good coordinate system) χ (upper triangular matrix containing the scaling factors and the calibrations). In a simplified way, we can for example consider that D is predetermined (it varies in practice very slowly) and that the only calibration parameter to be determined for the gyrometer 11 is b gyro , which actually tends to vary over time (we speak gyrometer 11).
    In a particularly preferred embodiment of double simultaneous calibration, there are only three calibration parameters to be determined: b gyro , A and b magnet0 . This limited number of parameters makes such a double calibration possible as soon as the object 1 has a “varied” trajectory (directions! Speeds which vary as a function of time).
    Magnetic equation
    The evolution of the magnetic field in the body frame (i.e. the frame of object 1) over time is described by the equation Μ = -ω x M + VM V + dM dt '
    The first term -ω x M describes the change in the field in the body frame which is due to the rotation of this frame with respect to the fixed frame. The second term VM V describes the change in the measured field resulting from the translation in a region with an inhomogeneous field. Finally, the third term takes into account the instarities of the magnetic field (for example periodic currents such as 50 Hz in Europe or movements of a magnet / a steel object near said object, etc.).
    If there is no "mobile" magnetic disturbance, i.e. the magnetic field has a stationary character, ~ 0, and we can reduce the equation to Μ = -ω x M + VM V.
    If there is no magnetic disturbance at all, i.e. the magnetic field has a uniform character in addition to stationary, VM "0 and we can even reduce the equation to Μ = -ω x M.
    This double assumption of uniformity and stationarity is normally most often verified. Indeed, for the Earth's magnetic field, uniformity is locally observed as a very good approximation in our latitudes, except in the case of the presence, for example, of a metallic structure or reinforced concrete (typically a bridge) in the surroundings. Instabilities due to displacements of objects in the vicinity are even rarer because the sources of disturbances would have to move, and possible instabilities due to periodic currents are on average zero can also treat it with estimation techniques, see for example document C.-l. Chesneau, M. Million, and C. Prieur, Motion estimation of a Rigid Body with an EKF using Magneto-Inertial Measurements, 7th Conf, on Indoor Positioning and Indoor Navigation (ΙΡΙΝΊ6), Madrid, Spain, 2016.
    So, with reference to FIG. 2, in step (c) we will try to calibrate the magnetometers 20 by making the double assumption of uniformity and stationarity of the field. The implementation of step (c) may be preceded by a step (b) of estimating a parameter representative of an error on the calibration parameters, step (c) being implemented if this parameter is greater than a predetermined threshold, ie shows that the magnetometers 20 need calibration.
    The method then advantageously comprises a possible estimation step (d) (first or second estimation depending on whether the method includes step (b)) of said parameter representative of an error on the calibration parameters by which the absence of disturbances, ie the relevance of the assumptions.
    We therefore understand that, in a double hypothesis of stationarity and uniformity of the magnetic field at the level of magnetometers 20, the quantity M ( estimate ^> + M ^ y ^ matl0n) x M ( estimate ) is equal to zero for calibration errors and near noise.
    Advantageously, the first expression (minimized in step (c)) is a function of | M < estimation ) + ω ^^ ηαίιοη - > x Under generic conditions, the minimization of these expressions will give unambiguous parameters for a generic trajectory . It will however be understood that alternatively the person skilled in the art can use as first / second expression any other functional sensitive to deviations from the relation defined by the first / second equation (standard L 2 , L œ , etc.).
    To implement this minimization, the data processing means 21 can work over time over an interval of a given length. As such, in a known manner a recursive filter (RLS methods, recursive least squares, etc.) or an optimization (least squares method, etc.) can be used.
    For example, assuming that the calibration parameters of the magnetometers 20 are determined simultaneously, ie the parameters A and bferdur, we can implement a principle called differential calibration in which we minimize ^ M (estlmatl0n > + x M (estlmatl0n) ^ dt with
    M ^ mation) = D (u ( ^ ure) + h) and M ^ estimate ^ = A- 1 (M ^ measure ï - b hard iron ) with the least squares.
    In another example, we can implement a principle called integral calibration. The idea is that after calibration, the measured field corresponds to the terrestrial field and must be constant in the terrestrial coordinate system (constant in norm and in direction), ie each component M = R b ^ t M ^ estimatl071 ^ = constant (field terrestrial) with / b ^ t the matrix of passage from the body reference to the terrestrial reference, determined from <^ gy S ro matlOn) We minimize the sum of the variances v (Mx) + var (M y ) + var ( M z ) (with var (M x ) = - ^% j (M x ^ -Μ *) 2 and mutatis mutandis for the other variances) to find the least squares calibration parameters. This method does not depend on the matrix / initial b ^ t .
    Error characterization
    As explained, the method advantageously comprises a step (b) and / or a step (d) of estimating a parameter representative of an error on the calibration parameters. As explained, in a preferred embodiment comprising both step (b) and step (d), step (c) is implemented if said parameter representative of an error estimated in step (b) is greater than a predetermined threshold, and step (d) is preferably implemented after each occurrence of step (c).
    The idea is to estimate the quality of the information provided by the magnetic 20 to rule out cases that are not favorable to calibration, i.e. cases of disturbances. As we will see, we can distinguish two possible causes for exceeding the threshold:
    i. an external (temporary) disturbance (eg a large steel / concrete structure next to which the vehicle is passing), which generates terms of type VM Ύ or γ in the magnetic equation.
    ii. a change in the magnetic properties of the car (e.g. change in the magnetization of the vehicle body, movement of a magnetic object by a passenger).
    The first occurrence of the estimation of the parameter representative of an error on the calibration parameters (step (b)) uses “original” calibration parameters, i.e. the parameters present at the time when the process is started. The second occurrence of the estimation of the parameter representative of an error on the calibration parameters (step (d)) uses the parameters as determined in step (c).
    If at the end of step (b) said parameter representative of an error is less than said predetermined threshold, it is known that both the original calibration parameters can be kept and that there is no magnetic disturbances.
    On the contrary, in the case of a parameter representative of an error greater than the threshold, we are in one of the cases i. and ii. previously mentioned. Then, the implementation of steps (c) and (d) makes it possible to distinguish these two cases.
    Thus, if at the end of step (d) said parameter representative of an error is still greater than said predetermined threshold, step (c) has not taken place under favorable conditions and the result of the calibration is not accepted. More precisely, and as shown in Figure 2, we can conclude that the problem is due to a temporary external disturbance (case i), and not to the calibration. We then maintain the original calibration for the attitude zone (see below) for the future (for example for when the external disturbance has been left behind), i.e. we keep the parameters of the old calibration as calibration. Magnetic measurements can possibly be declared unusable as long as the problem persists.
    On the contrary, if said parameter representative of an error is now lower than said predetermined threshold, it is because we were in case ii, ie the cause of the problem was the original calibration, and step (d) comprises the calibration of magnetometers 20 and / or inertial measurement means 11 with the values determined in step (c) of the calibration parameters.
    Note that determined values of the calibration parameters during an occurrence of step (c), but not used for the actual calibration, can be stored on the data storage means 12, and used during a future occurrence of step (d). For example, we can predict that as long as the parameter representative of an error is above the threshold, we store the determined calibration parameters, and when we go below the threshold, the actual calibration also takes into account the stored values .
    In general, said parameter representative of an error in the calibration parameters is a function of the value of said expression to be minimized, calculated for the determined values of the calibration parameters.
    A first embodiment of step (b) / (d) is said to be intrinsic since it uses only quantities available in step (c). Preferably, the estimation residues of step (c) are used, ie said parameter representative of an error is in particular the norm (eg L 2 or L 00 ) of the value of the first / second expression over a given time interval, typically ^ M (estlmatl0n > + x M (estlmatl0n) ^ dt. In such a case, step (d) can be implemented concomitantly with step (c) In the case of a recursive filter, we can use the norm (eg L 2 or L 00 ) of the innovation of the filter over a given period of time. Note that in the case of the use of residues for step (b), any set of calibration parameters and any set of acquired data can be used - the parameters need not be calculated using the current acquired data (obtained during the last implementation of step (a)). Thus, we can characterize before (or even without) implementing step (c).
    Alternatively, we can quantify the deviation from a normal distribution, ie see if the statistical distribution of the values of ^ (estimate) _ | _ ω ^ 77ηαί10η ^ (estimate) θθ | auß d in g or n O n, p ar example ΘΠ calculating the higher moments.
    According to a second mode, one can calculate the norm of the estimated magnetic field, and compare it with a threshold: if the estimated magnetic field is abnormally high compared to the reference value of the terrestrial field, it is that there is a magnetic disturbance.
    According to a third embodiment, one can simply estimate a spatial gradient of the components of the magnetic field, i.e. VM.
    Indeed, this estimate must be close to zero for the uniformity to be retained. It is desirable in this mode to have three magnetometric thaxes, and in any event at least eight magnetometers 20, to determine the coefficients of the field as well as its gradient.
    Alternatively or in addition, we can use learning to improve the estimation of this error parameter and / or develop an approach to identify favorable conditions (ie stationary and uniform) in a more robust manner and with availability. increased.
    In particular, we can implement learning mechanisms such as neural networks, support vector machines, nearest neighbor methods, decision tree forests, etc. Thus, at each occurrence of steps (b) to (d), we can enrich a learning base in which each set of measurement data is "tagged" with the corresponding value of the parameter representative of an error, so to progressively (as successive occurrences of steps (b) to (d)) and automatically learn to distinguish acceptable calibrations from those that are not acceptable. Thus, the calibration continuously improves itself.
    Calibration area
    In theory, there is only one set of calibration parameters that corrects the defects in the measurement of the magnetic field for all of the attitudes of object 1.
    In practice, along a typical vehicle journey for a limited period of time, only a subset of the attitudes are explored. Optimization for this attitude subset produces a set of calibration parameters which results from local minimization (on the attitude subset considered). It may therefore be that this set of calibration parameters is not optimal outside the attitude subspace considered, which is called “calibration zone”.
    Preferably, the method further comprises, if said parameter representative of an error is less than a predetermined threshold, a step (e) of determining a subset of the attitudes of the object for which the calibration is relevant. . Note that there may be several subsets used for several calibrations.
    This can significantly improve the calibration: during the movement of object 1, we can distinguish two cases according to the current attitude:
    i. If the attitude of object 1 is not covered by any relevant subset, we can implement the present method or otherwise choose a default calibration in an arbitrary manner.
    ii. If the attitude of the current object 1 is covered by a relevant subset, the appropriate calibration for this zone is used.
    Said subset can be characterized for example by intervals of Euler angles, or by another parametrization of the variety of attitudes (quaternions, orthogonal matrices).
    With regard to its determination, it can be determined by the convex envelope of the attitude data from all of the measurement points, or by the set of measurements itself.
    Motion estimation
    As explained, the method advantageously comprises a step (f) of estimation by the data processing means 21 of the movement of said object 1 as a function of the angular speed of the object 1 and / or of the components of the magnetic field, and / or of a possible measured linear speed of the object 1 (by the means 10), and of the values of the calibration parameters, ie after recalibration. This step (f), which is not shown in FIG. 2 which focuses on the calibration, can be carried out continuously in parallel.
    By motion estimation, we mean in particular at least the estimation of an orientation of the object 1 (in the horizontal plane, i.e. a heading) and advantageously the estimation of a speed standard. We can either simply implement navigation with the magnetic heading (ie with a "compass") obtained with magnetometric measurements, or gyrometric navigation or even a magneto-gyrometric fusion in the case of double calibration. In the latter case, the orientation is typically obtained by integrating the angular velocity.
    If necessary, the linear speed measured (by any means 10) is only used in step (f) to determine an overall speed of the object 1.
    Step (f) can also comprise the calculation as a function of said parameter representative of an error on the calibration parameters, of a magnetometric or gyro-magnetometric error in orientation (heading). For example, the heading error accumulated during a period after a calibration can be estimated by the uncertainty in the estimate of the bias multiplied with the duration of this period.
    Note that in the case of an autonomous vehicle, step (f) may comprise the generation of a command from said vehicle 1 as a function of the estimated movement, so as to bring the vehicle 1 for example to a desired destination, or to stop the vehicle 1 while keeping it in a path devoid of obstacles.
    Equipment and object
    According to a second aspect, the invention relates in particular to a set of equipment 11, 20, 21 and possibly 10 for the implementation of one or the other of the embodiments of the method.
    This set can be installed as a kit in a “classic” object 1 so as to transform it. Alternatively, in particular in the case where the object 1 is a vehicle, it may be an autonomous vehicle already provided with data processing means 21 for the navigation of the vehicle, as well as sensors such as the gyrometer 11 and / or odometers serving as additional acquisition means 10.
    In particular, an object 1 is proposed, in particular of the wheeled vehicle type, comprising:
    - inertial measurement means 11 (for example gyrometer) configured to acquire an angular speed of the object 1;
    - magnetometers configured to acquire at least three components of the magnetic field if necessary,
    - possibly additional acquisition means 10 configured to acquire a measured linear speed of the object 1 (advantageously odometers fitted to at least two of the vehicle wheels and configured to acquire measured speeds of said two wheels);
    - data processing means 21 configured to determine values of at least one calibration parameter of the magnetometers 20 minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation on the angular speed of the object 1, the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers 20, and the at least one magnetic equation assuming a uniform and stationary character of the magnetic field at the level of the magnetic measurement means 20 .
    - the data processing means 21 which can also be configured to o estimate a parameter representative of an error on the calibration parameters before and / or after the determination of the values of the calibration parameters;
    o estimate a movement of said object 1.
    As explained above, the object 1 can also include a memory 22 and an interface 23.
    Computer program product
    According to a third and a fourth aspect, the invention relates to a computer program product comprising code instructions for the execution (on the processing means 21) of a method for calibrating the 10 magnetometers according to the first aspect of the invention, as well as storage means readable by computer equipment (for example data storage means 22) on which is found this computer program product.
    权利要求:
    Claims (15)
    [1" id="c-fr-0001]
    1. Method for calibrating magnetometers (20) equipping an object (1) evolving in an ambient magnetic field, the method being characterized in that it comprises steps of:
    (a) Acquisition
    - by the magnetometers (20), at least three measured components of the magnetic field at the level of the magnetometers (20), and
    - By inertial measurement means (11) integral with said object (1), with an angular speed of the object (1);
    (c) Determination by data processing means (21) of values of at least one calibration parameter of the magnetometers (20) minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation over the angular velocity object (1),
    the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers (20), and
    - at least one magnetic equation assuming a uniform and stationary nature of the magnetic field at the level of the magnetic measurement means (20).
    [2" id="c-fr-0002]
    2. Method according to claim 1, in which the estimated components M (estimate ) of the magnetic field are linked to the measured components M (measurement ) by a model M ( measurement ) = a {^ (estimate) + b magnet0 , where A and b magnet0 are the calibration parameters of the magnetometers (20).
    [3" id="c-fr-0003]
    3. Method according to one of claims 1 and 2, further comprising a step (b) of estimating a parameter representative of an error on the calibration parameters, step (c) being implemented if said parameter representative of an error is greater than a predetermined threshold.
    [4" id="c-fr-0004]
    4. Method according to claim 3, further comprising a step (d) of new estimation of said parameter representative of an error on the calibration parameters so as to distinguish an external magnetic disturbance from a change in magnetic properties of the object (1).
    [5" id="c-fr-0005]
    5. The method of claim 4, further comprising, if at the end of step (d) said parameter representative of an error is less than a predetermined threshold, a step (e) of determining a subset of the attitudes of the object (1) for which the calibration is relevant.
    [6" id="c-fr-0006]
    6. Method according to one of claims 1 to 5, wherein step (c) comprises the implementation of a recursive filter or an optimization.
    [7" id="c-fr-0007]
    7. Method according to one of claims 1 to 6, in which the inertial measurement means (11) are a gyrometer, the angular speed of the object (1) acquired in step (a) being a measured angular speed, and that used by the magnetic equation or equations is an angular speed estimated as a function of the angular speed measured and of calibration parameters of the gyrometer (11), step (c) also comprising the determination of values of at least one parameter for calibrating the gyrometer (11).
    [8" id="c-fr-0008]
    8. The method as claimed in claim 7, in which the estimated angular speed œ ^ mation) of the object (1) is linked to the angular speed measured wj ™ e o sure) by a model œ ^ mation} = D (œ ^ ure} + bgyro)> ° ù D and bgyro are the gyrometer calibration parameters (11).
    [9" id="c-fr-0009]
    9. Method according to one of claims 1 to 8, wherein the magnetic equation is of the form M - -ωχΜ, where M is the vector of the components of the magnetic field, and ω the angular velocity.
    [10" id="c-fr-0010]
    10. The method of claim 9, wherein said expression is a function of | M ( estMnatIon ) + χ M (estimate) |
    [11" id="c-fr-0011]
    11. The method according to claims 3 and 10 in combination, wherein said parameter representative of an error in the calibration parameters is either the mean of | Af ( estl ™ atlon ) + x M (esti, nation q over an interval of given time, ie a spatial gradient of said components of the magnetic field.
    [12" id="c-fr-0012]
    12. Method according to one of claims 1 to 11, comprising a step (f) of estimation by the data processing means (21) of the movement of said object (1) as a function of the angular speed of the object ( 1), of the measured components of the magnetic field, and of the values of the calibration parameters.
    [13" id="c-fr-0013]
    13. Object (1) evolving in an ambient magnetic field, comprising inertial measurement means (11) configured to acquire an angular speed of the object (1), magnetometers (20) configured to acquire at least three components of the field magnetic, the object (1) being characterized in that it further comprises data processing means (21) configured for:
    - determining values of at least one magnetometer calibration parameter (20) minimizing an expression defined by estimated components of the magnetic field and at least one magnetic equation on the angular speed of the object (1),
    the estimated components of the magnetic field being a function of the measured components of the magnetic field and of calibration parameters of the magnetometers (20), and
    - at least one magnetic equation assuming a uniform and stationary nature of the magnetic field at the level of the magnetic measurement means (20).
    [14" id="c-fr-0014]
    14. A computer program product comprising code instructions for the execution of a magnetometer calibration method (20) according to one of claims 1 to 12, when said program is executed on a computer.
    [15" id="c-fr-0015]
    15. Storage means readable by computer equipment on which a computer program product comprises code instructions for the execution of a magnetometer calibration method (20) according to one of claims 1 to 12.
    类似技术:
    公开号 | 公开日 | 专利标题
    WO2019239062A1|2019-12-19|Method for calibrating magnetometers fitted in an object
    EP3213033B1|2018-08-22|Method of estimating a navigation state constrained in terms of observability
    FR2886501A1|2006-12-01|METHOD AND DEVICE FOR LOCALIZING A TERMINAL IN A WIRELESS LOCAL NETWORK
    EP3658921B1|2021-04-21|Method for calibrating a magnetometer
    FR2906881A1|2008-04-11|METHOD FOR FUNCTIONALLY CONTROLLING AN INERTIAL POWER PLANT OF A MOBILE.
    FR2927418A1|2009-08-14|METHOD AND SYSTEM FOR VALIDING AN INERTIAL POWER PLANT OF A MOBILE.
    EP2573515B1|2019-11-06|Vibrating gyroscope calibration
    EP3447654A1|2019-02-27|Method for determining the trajectory of a moving object, program and device for implementing said method
    EP3004807B1|2016-12-28|Method for autonomous calibration of an inertial equipment used in a static mode
    WO2009083374A1|2009-07-09|Method for independent alignment of an inertial unit for an onboard instrument to be mounted e.g. in an aircraft, and onboard instrument using such method
    EP3655724B1|2021-05-05|Method for estimating the movement of an object moving in a magnetic field
    FR3082612A1|2019-12-20|CALIBRATION METHOD OF A GYROMETER EQUIPPED WITH AN OBJECT
    EP3655800B1|2021-09-01|Method and device for magnetic field measurement by magnetometers
    EP2410293B1|2019-11-06|Method and system for harmonising an angular positioner reference in relation to an Earth-based reference frame
    FR2994258A1|2014-02-07|METHOD FOR COMPENSATING A SENSOR SIGNAL OF A VEHICLE AND INFORMATION SYSTEM APPLYING SUCH A METHOD
    FR3067138A1|2018-12-07|METHOD FOR ESTIMATING THE POSITION OF A MAGNET COMPRISING AN IDENTIFICATION PHASE OF A MAGNETIC DISTURBATOR
    FR3069633B1|2019-08-23|CAP DETERMINATION FROM THE MEASURED FIELD BY MAGNETIC SENSORS
    EP3623758B1|2021-04-21|Positioning system, and associated method for positioning
    FR3097316A1|2020-12-18|Method for monitoring the performance of inertial measurement units
    EP3899432A1|2021-10-27|Method for characterising an inertial measurement unit
    EP3655725A1|2020-05-27|Method for estimating the movement of an object moving in a magnetic field environment
    FR3109991A1|2021-11-12|Method of estimating a physical quantity
    FR3050524A1|2017-10-27|METHOD FOR ESTIMATING THE MOVEMENT OF A EVOLVING OBJECT IN A MAGNETIC FIELD
    同族专利:
    公开号 | 公开日
    JP2021527214A|2021-10-11|
    FR3082611B1|2020-10-16|
    US11248932B2|2022-02-15|
    EP3807594A1|2021-04-21|
    WO2019239062A1|2019-12-19|
    CN112334736A|2021-02-05|
    US20210247206A1|2021-08-12|
    KR20210021035A|2021-02-24|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20020100178A1|2000-12-06|2002-08-01|Honeywell International Inc.|Electronic compass and compensation of large magnetic errors for operation over all orientations|
    US20110178707A1|2010-01-21|2011-07-21|Invensense, Inc.|Apparatus and methodology for calibration of a gyroscope and a compass included in a handheld device|
    US20120203486A1|2011-02-08|2012-08-09|Research In Motion Limited|System and Method for Calibrating an Accelerometer|CN112964278A|2021-03-25|2021-06-15|北京三快在线科技有限公司|Method, device, electronic equipment and storage medium for determining calibration parameters of magnetometer|FR2728338B1|1994-12-19|1997-02-28|
    US6860023B2|2002-12-30|2005-03-01|Honeywell International Inc.|Methods and apparatus for automatic magnetic compensation|
    US7451549B1|2006-08-09|2008-11-18|Pni Corporation|Automatic calibration of a three-axis magnetic compass|
    FR2914739B1|2007-04-03|2009-07-17|David Jean Vissiere|SYSTEM PROVIDING THE SPEED AND POSITION OF A BODY USING MAGNETIC FIELD VARIATIONS EVALUATED THROUGH MEASUREMENTS OF MAGNETIOMETERS AND ONE OR MORE INERTIAL PLANTS|
    US8531180B2|2010-03-30|2013-09-10|Apple Inc.|Determining heading using magnetometer data and angular rate data|
    US8717009B2|2010-10-06|2014-05-06|Apple Inc.|Magnetometer calibration|
    EP2638362A4|2010-11-08|2014-03-26|Alpinereplay Inc|Device and method of gyro sensor calibration|
    US20130245984A1|2010-11-17|2013-09-19|Hillcrest Laboratories, Inc.|Apparatuses and methods for magnetometer alignment calibration without prior knowledge of the local magnetic field|
    WO2012135960A1|2011-04-06|2012-10-11|Uti Limited Partnership|System and method for gyroscope error estimation|
    US20130018582A1|2011-07-13|2013-01-17|Miller Paul A|Inertial Navigation Common Azimuth Reference Determination System and Method|
    JP2013064695A|2011-09-20|2013-04-11|Yamaha Corp|State estimating device, offset updating method, and offset updating program|
    WO2013125242A1|2012-02-24|2013-08-29|旭化成株式会社|Offset estimation device, offset estimation method, offset estimation program and information processing device|
    JP2014089113A|2012-10-30|2014-05-15|Yamaha Corp|Posture estimation device and program|
    US8915116B2|2013-01-23|2014-12-23|Freescale Semiconductor, Inc.|Systems and method for gyroscope calibration|
    US9417091B2|2013-05-13|2016-08-16|The Johns Hopkins University|System and method for determining and correcting field sensors errors|
    US10175043B2|2013-05-15|2019-01-08|FLIR Belgium BVBA|Toroidal shape recognition for automatic compass calibration systems and methods|
    US9939532B2|2015-01-09|2018-04-10|Honeywell International Inc.|Heading for a hybrid navigation solution based on magnetically calibrated measurements|
    JP6705972B2|2016-05-20|2020-06-03|サイトセンシング株式会社|Attitude estimation device, attitude estimation method, control program, and recording medium|
    US10663298B2|2017-06-25|2020-05-26|Invensense, Inc.|Method and apparatus for characterizing platform motion|
    FR3069649B1|2017-07-26|2021-01-01|Sysnav|CALIBRATION PROCESS OF A MAGNETOMETER|
    CN108051761A|2017-09-06|2018-05-18|哈尔滨工程大学|A kind of three axle magnetometer itself error on-line correction method|
    WO2019081747A1|2017-10-26|2019-05-02|Sony Semiconductor Solutions Corporation|Heading determination device and method, rendering device and method|
    US10883830B2|2019-01-15|2021-01-05|Lockheed Martin Corporation|Terrestrial land air and sea navigation system|CN112461224B|2020-11-10|2021-09-14|武汉大学|Magnetometer calibration method based on known attitude angle|
    CN112762965B|2021-04-08|2021-09-07|北京三快在线科技有限公司|Magnetometer calibration method and device|
    法律状态:
    2019-06-12| PLFP| Fee payment|Year of fee payment: 2 |
    2019-12-20| PLSC| Publication of the preliminary search report|Effective date: 20191220 |
    2020-06-11| PLFP| Fee payment|Year of fee payment: 3 |
    2021-05-27| PLFP| Fee payment|Year of fee payment: 4 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1855160|2018-06-13|
    FR1855160A|FR3082611B1|2018-06-13|2018-06-13|METHOD OF CALIBRATION OF MAGNETOMETERS EQUIPPING AN OBJECT|FR1855160A| FR3082611B1|2018-06-13|2018-06-13|METHOD OF CALIBRATION OF MAGNETOMETERS EQUIPPING AN OBJECT|
    CN201980039944.0A| CN112334736A|2018-06-13|2019-06-12|Method for calibrating a magnetometer of an object|
    PCT/FR2019/051414| WO2019239062A1|2018-06-13|2019-06-12|Method for calibrating magnetometers fitted in an object|
    EP19745679.1A| EP3807594A1|2018-06-13|2019-06-12|Method for calibrating magnetometers fitted in an object|
    KR1020217001141A| KR20210021035A|2018-06-13|2019-06-12|How to calibrate a magnetometer fitted to an object|
    JP2020569029A| JP2021527214A|2018-06-13|2019-06-12|How to calibrate an object's magnetometer|
    US16/973,950| US11248932B2|2018-06-13|2019-06-12|Method for calibrating magnetometers of an object|
    [返回顶部]