• 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:
    This method of locating magnetic objects comprises: - the identification (168; 211), as a function of the moving objects presented in front of a network of magnetometers, of an additional equation connecting, by a relation of equality, a first variable of 'a current system of one-term equations, that term being a predefined relation between one or more other variables of the current system of equations or a constant numerical value, - the replacement (180; 216), in the current system of equations, from this first variable by the term to which it is equal to obtain a new system of equations in which the number of variables whose value is to be estimated is smaller than in the current system of equations, - the use of the new system of equations instead of the current system of equations for estimating the position or orientation or amplitude of a magnetic moment of one of the magnetic objects.
    公开号:FR3015049A1
    申请号:FR1362730
    申请日:2013-12-16
    公开日:2015-06-19
    发明作者:Tristan Hautson
    申请人:Commissariat a lEnergie Atomique CEA；Commissariat a lEnergie Atomique et aux Energies Alternatives CEA；
    IPC主号:
    专利说明:

    [0001] FIELD OF THE INVENTION The invention relates to a method and a device for locating mobile magnetic objects presented in front of a network of magnetometers.  The invention also relates to an information recording medium for the implementation of this method.  [002] Patent applications FR 2 988 862 and FR 2 988 874 disclose methods for locating mobile magnetic objects presented in front of a magnetometers network comprising N tri-axis magnetometers mechanically connected to each other without any degree of freedom to maintain a known distance between each of these magnetometers, where N is an integer greater than or equal to five.  These known methods comprise: a) the measurement by each of the magnetometers of the amplitude of the magnetic field along each of its measurement axes, b) the estimation from the measurements of the magnetometers, the values of several variables by solving a current system of equations connecting these variables to each measurement of a tri-axis magnetometer of the array, each variable corresponding to the position or orientation or amplitude of a magnetic moment of a permanent magnet of the magnetic object, - the iteration of steps a) and b) at successive instants in time to obtain the estimated values of each variable at these different successive instants.  [003] These methods work particularly well.  In particular, these methods are capable of locating many different moving magnetic objects.  In addition, they are also able to simultaneously locate several moving magnetic objects mechanically independent of each other.  Because of these capabilities, these processes are said to be "flexible".  [004] However, when the number of mobile magnetic objects to be simultaneously located increases, the accuracy of the location of each of these mobile magnetic objects decreases.  [005] The invention aims to improve the accuracy of these known methods for locating mobile magnetic objects, especially when the number of mobile magnetic objects to be simultaneously located increases, while maintaining their flexibility.  [006] It therefore relates to such a method comprising: c) identifying, as a function of the moving objects presented in front of the magnetometer array, an additional equation connecting, by a relation of equality, a first variable of the current system of equations with a term, this term being a predefined relation between one or more other variables of the current system of equations or a constant numerical value, d) the replacement, in the current system of equations, of this first variable by the term to which it is equal, identified in step c), to obtain a new system of equations in which the number of variables whose value is to be estimated is smaller than in the current system of equations, e) the use of the new system of equations instead of the current system of equations during the following iterations of steps a) and b), the new system of equations thus becoming, for the iterations following steps a) and b), the present system of equations.  [7] The Applicant has found that reducing the number of variables in the system of equations used in step b) makes it possible to reduce the estimation noise on the estimated positions and orientations of the moving magnetic objects.  This therefore results in an improvement in the location accuracy of the moving magnetic objects.  However, this also reduces the flexibility of the process.  [8] To reduce this number of variables without impairing the flexibility of the process, the above method begins by using a system of equations having many independent variables.  Next, based on the moving objects presented in front of the magnetometer array, it identifies additional equations linking together variables of the current system of equations or connecting a variable of the current system of equations to a constant value.  These additional equations are then used to reduce the number of variables in the current system of equations.  Thus, this method automatically and dynamically adapts the number of variables of the system of equations as a function of the mobile magnetic objects to be located.  This makes it possible to maintain the flexibility of the process while improving the accuracy of location.  [9] In addition, when the number of variables for which a value is to be estimated in step b) is reduced, the location method is faster.  [0010] The above method does not require storing a large number of predefined equation systems but simply a system of initial equations and additional equations.  Finally, reducing the number of variables to be estimated also increases the robustness of the process.  Embodiments of this method may include one or more of the following features: the identification of the additional equation is performed according to the estimated values, during a previous iteration of steps a) and b) variables; the identification of the additional equation comprises: calculating a magnitude representative of the amplitude of the variations of the first variable during previous iterations of steps a) and b) from the values of this first estimated variable; during these previous iterations of steps a) and b), the comparison of this calculated quantity with a predetermined threshold to establish if this first variable has varied during these previous iterations of steps a) and b), and - if this predetermined threshold is crossed, the establishment of a new additional equation connecting, by a relation of equality, the value of this first variable to a constant value, the constant value being a function of at least one value of this first variable estimated during these previous iterations of steps a) and b) during which it has been established that its value does not vary, then the use of this new equation. it is added during step d), and - if this predetermined threshold is not crossed, the absence of establishment of this new additional equation; step c) comprises: acquiring a distinctive characteristic of the mobile magnetic object presently presented in front of the magnetometer array, and comparing the acquired distinctive characteristic with pre-recorded selection conditions in a base of data, this database associating with each selection condition at least one additional equation linking, by an equality relation, a variable of the current system of equations to a term, this term being a predefined relationship between one or more other variables of the current system of equations or a constant numerical value, and - only if the distinctive characteristic acquired corresponds to one of the pre-recorded selection conditions, then the use in step d) of the additional equation associated with that condition of selection; the acquisition of a distinguishing characteristic involves obtaining this distinctive characteristic from the values of the variables estimated using the present system of equations; the method comprises: calculating an estimation error with the current system of equations, the estimation error being representative of the difference between: estimated values of the measurements of the magnetometers when the position, the orientation and amplitude of the magnetic moment of each magnetic object are equal to those estimated using the current system of equations during an iteration of steps a) and b), and the measurements of the magnetometers recorded in step a) during the same iteration of steps a) and b), and if the estimation error calculated crosses a predetermined threshold: the selection of at least one of the additional equations previously used in step d) to obtain the the current system of equations, replacing, in the current system of equations, the term of the selected supplementary equation by the variable that this term has replaced to obtain a new system of equations in which the number e of variables whose value is to be estimated is greater than in the current system of equations and the use of the new system of equations instead of the current system of equations during the following iterations of steps a) and b) and, if the calculated error does not exceed this predetermined threshold, maintaining the use of the current system of equations during the following iterations of steps a) and b); the selection of at least one additional equation comprises: selecting at least two additional equations previously used in step d) to obtain the current system of equations; for each of the additional equations selected, replacing in the current system of equations, of the term of the selected supplementary equation, by the variable that this term has replaced to obtain third is fourth system of different equations, each of these fourth and third systems of equations containing a number of variables whose value must be estimated to be greater than in the current system of equations, - the calculation of the estimation error with these third and fourth systems of equations, and - the selection of the additional equation, among those initially selected, which minimizes the calculated estimation error.  These embodiments of the location method further have the following advantages: using the previously estimated values to identify the additional equation allows the use of the magnetometers both for locating the magnetic objects and for selecting the additional equation, 35 - in the absence of variation of one of the variables, establishing the additional equation according to which this variable is equal to a constant value makes it possible to reduce the number of variables to be estimated without using pre-recorded knowledge on the possible behavior (s) of this variable, - using a threshold to determine if a variable varies, this threshold being determined from the maximum variation observed for this variable, makes it possible to increase the precision of the process, - the identification of an additional equation by comparing the distinguishing characteristics of a moving object to condi pre-recorded selection allows to limit the number of variables to estimate by taking into account a predefined knowledge base on the possible displacements of moving magnetic objects, - to obtain the distinctive characteristic from the measurements made by the network of 10 magnetometers allows to identify the additional equation using the same magnetometers as those used to locate the moving magnetic object; - adding again a variable to the current system of equations when the estimation error exceeds a certain threshold makes it possible to automatically adapt the process to a change of mobile magnetic object or to a change in its use; 15 - select the variable to reintroduce into the current system of equations from the estimation error calculated in the case where this variable is absent from the system of equations and in the case where this variable is present in the system of equations equations allows to systematically reduce the error each time a variable is reintroduced into the system of equations.  The invention also relates to an information recording medium comprising instructions for the execution of the above method, when these instructions are executed by an electronic computer.  Finally, the invention also relates to a device for locating mobile magnetic objects, this device comprising: a magnetometers network comprising N tri-axis magnetometers mechanically connected to each other without any degree of freedom for maintain a known distance between each of these magnetometers, where N is an integer greater than or equal to five, - a processing unit capable of: a) acquiring the measurements of each of the magnetometers of the magnetometer array, b) estimating, from the measurements of the magnetometers, values of several variables by solving a current system of equations connecting these variables to each measurement of a tri-axis magnetometer of the network, each variable 35 corresponding to the position or the orientation or the amplitude of a magnetic moment of a magnetic object, - to repeat steps a) and b) at successive instants in time to obtain the estimated values of each variable at these different successive instants, in which the processing unit is also able: c) to identify, according to the moving objects presented in front of the magnetometers network, an additional equation connecting, by a relation of equality, a first variable of the current system of equations to a term, this term being a predefined relation between one or more other variables of the current system of equations or a constant numerical value, d) to replace, in the current system of equations, this first variable by the term to which it is equal, identified in step c), to obtain a new system of equations in which the number of variables whose value is to be estimated is smaller than in the current system of equations e) to use the new system of equations instead of the current system of equations 15 during the following iterations of steps a) and b), the new equation system. Thus, for the following iterations of steps a) and b), the current system of equations becomes.  The invention will be better understood on reading the description which will follow, given solely by way of nonlimiting example and with reference to the drawings, in which: FIG. 1 is a schematic illustration of a man-machine interface for controlling an electrical apparatus.  FIG. 2 is a partial illustration, in plan view, of an array of magnetometers used in the human-machine interface of FIG. 1; FIG. 3 is a diagrammatic illustration, in vertical section, of a cursor that can be used in the interface of FIG. 1; FIG. 4 is a schematic illustration, in vertical section, of a button that can be used in the interface of FIG. 1; Figure 5 is a schematic illustration of a database used in the interface of Figure 1; FIG. 6 is a flowchart of a method of controlling an electrical apparatus using the human-machine interface of FIG. 1.  In these figures, the same references are used to designate the same elements.  In the rest of this description, the features and functions well known to those skilled in the art are not described in detail.  FIG. 1 represents a man-machine interface 2 for controlling an electrical apparatus 4.  Here, the electrical apparatus comprises a screen and a control unit capable of controlling the display of an image on this screen.  Here, the operation of the interface 2 is illustrated in the case where the unit 5 is a video game console.  For example, unit 5 controls the movement of a character 6 on the screen.  However, the interface 2 can be used in many other applications as described at the end of this description.  The interface 2 comprises several utensils operable directly by hand by a human, referred to as "user" thereafter.  Each of these utensils has at least one magnetic object.  Here each magnetic object is a permanent magnet.  For simplicity of Figure 1, only a utensil 10 is shown in this figure.  Other utensils of the interface 2 are described with reference to FIGS. 3 and 4.  The interface 2 also comprises a device 12 for locating each permanent magnet.  In this embodiment, to modify the configuration of the interface 2, each utensil is freely movable, directly by the hand of the user, in an orthogonal reference Rm fixed without any degree of freedom to the device 12.  Here, the X and Y directions of the reference Rm are horizontal and the direction Z is vertical.  For this purpose, each utensil weighs less than one kilo and, preferably, less than 200g.  The dimensions of each utensil are sufficiently small that it can be grasped and moved by one hand of the user.  In this embodiment, the utensil 10 comprises a handle 14, a permanent magnet 16 and a support 18.  The handle 14 has an oblong shape to form a handle easily grasped by the user.  It extends along a longitudinal axis 17.  The joystick 14 is intended to be used as a joystick.  For this purpose, a lower end of the handle 14 is fixed to the support 18 by means of a ball joint 19.  The support 18 is provided with a lower flat face intended to be fixed or to rest on a plane so as to immobilize, in the frame Rm, the position of the ball 19 when the utensil 10 is used.  Thus, the lever 14 can be freely inclined by the user around the center of rotation of the ball 19.  In use, the lever 14 thus has three degrees of freedom in rotation.  For example, the lever 14, the support 18 and the ball 19 are entirely made of a non-magnetic material, that is to say a material having no measurable magnetic property by the device 12.  This material is, for example, plastic.  The position of the handle 14 around the X, Y and Z axes is located from the position of the magnet 16.  The magnet 16 has a non-zero magnetic moment even in the absence of an external magnetic field. Typically, in this description, the coercive magnetic field of each magnet is greater than 100 A. m-1 or 500 A. m-1.  For example, it is made of ferro- or ferrimagnetic material.  The power of each permanent magnet is typically greater than 0.01 A. m2 or 0.1 A. m2.  The magnet 16 is fixed without any degree of freedom on the handle 14.  The direction of the magnetic moment of the magnet 16 here coincides with the longitudinal axis 17 of the lever 14.  In Figure 1 and the following, the direction of the magnetic moment of a magnet is represented by an arrow.  The greatest length of this magnet is noted L thereafter.  When using the joystick 14, the coordinates of the magnet 16 are connected to each other by the following vector relationship: P + Rd * (M / M II) = Po, where - Po is the vector encoding the position of the center of rotation of the ball 19 of the utensil 10 in the reference Rm, - P is the vector encoding the position of the geometric center of the magnet 16 in the reference Rm, 15 - M is the moment magnet magnet 16, - II M II is the amplitude of the magnetic moment M, - Rd is the distance that separates the center of rotation of the ball 19 from the geometric center of the magnet 16.  From this vectorial relationship, it is possible to establish three additional equations 20 interconnecting the variables coding the position, the orientation and the amplitude of the magnetic moment of the magnet 16.  For example, these additional equations are as follows: / m × 2 + m-y 2 + m-z 2 Rd * Mx = Px-Pox (1) 25 -Rd * My = Py-Poy (2) ~ IMx 2 + M y2 + m z2 Rd * Mz = Pz-Poz (3), Liv y2 + m-z2 where: 30 - Mx, My and Mz are the orthogonal projections of the magnetic moment M on the X, Y and Z axes of the reference mark Rm .  These equations express the fact that: - the geometric center of the magnet 16 is always located on a sphere centered on the center of rotation of the ball 19 and radius Rd, and - that the magnetic moment of the magnet 16 points permanently on the center of rotation of the ball 19.  The center of geometry of an object is the center of gravity of all the points of this object by assigning the same weight to each of these points.  To simplify the description, in the following description, the additional equations are expressed in a local coordinate system without any degree of freedom to the utensil.  It is within the abilities of those skilled in the art to express these same additional equations in the reference Rm taking into account the coordinates and the orientation of the local coordinate system relative to the reference Rm.  It is simply a change of reference.  The device 12 makes it possible to locate one or more magnets in the reference frame Rm.  By locating a magnet, here is meant the determination of the position and the orientation of the magnet 16 in the reference Rm.  The position is unambiguously defined by the values of three variables, for example the x, y and z coordinates in the Rm frame.  More precisely, in the case of a magnet, the variables x, y and z are the coordinates of the geometric center of the magnet.  The orientation of the magnetic moment of a magnet is defined in the frame Rm by the values of two variables ey and 0 ,.  Here, the variables ey and 0, are the angles of the magnetic moment of the magnet, respectively, with respect to the Y and Z axes of the mark Rm.  The device 12 also determines a sixth variable A.  Variable A is the magnitude of the magnetic moment of the magnet.  The device 12 comprises for this purpose a network of N tri-axis magnetometers Mi.  In Figure 1, the vertical wavy lines indicate that a portion of the device 12 has not been shown.  Typically, N is greater than five and preferably greater than sixteen or thirty-two.  Here, N is greater than or equal to sixty-four.  In this embodiment, the magnetometers Mi; are aligned in rows and columns to form a matrix.  Here, this matrix has eight rows and eight columns.  The indices i and j respectively identify the row and the column of this matrix at the intersection of which the magnetometer ni is located.  In Figure 1, only the magnetometers Mil, -M2, -Mi3, -Ni Mi8 of a line i are visible.  Position - Mi magnetometers; relative to each other is described in more detail with reference to FIG.  Each magnetometer Mi; is fixed without any degree of freedom to the other 35 magnetometers.  For this purpose, the magnetometers Mi; are fixed without any degree of freedom on a rear face 22 of a rigid plate 20.  This rigid plate has a front face 24 turned towards the magnet 16.  The plate 20 is made of a rigid non-magnetic material.  For example, the plate 20 is made of glass.  Each magnetometer Mu measures the direction and intensity of the magnetic field generated by the magnets present in front of the face 24.  For this purpose, each magnetometer Mu measures the standard of the orthogonal projection of the magnetic field at the level of this magnetometer Mu on three measurement axes of this magnetometer.  Here, these three measurement axes are orthogonal to each other.  For example, the measurement axes of each of the magnetometers Mu are, respectively, parallel to the X, Y and Z axes of the marker.  The sensitivity of the magnetometer Mu is for example 4 * 10-7T.  Each magnetometer Mu is connected via an information transmission bus 28 to a processing unit 30.  The processing unit 30 is capable, for each magnet, to locate it in the reference Rm and to estimate the amplitude of its magnetic moment from the measurements of the magnetometers Mu.  For this purpose, the unit 30 comprises a programmable electronic computer 32 capable of executing instructions recorded on an information recording medium.  The unit 30 thus also comprises a memory 34 containing the instructions necessary for the execution by the computer 32 of the method of FIG. 6.  In this embodiment, for each number P of magnetic objects capable of being simultaneously used in the interface 2, the unit 30 implements a mathematical model Mp associating each measurement of a magnetometer Mu with the positions, orientations and amplitudes. magnetic moments of P magnetic objects in the reference Rm.  Each model M p is in the form of a system of equations in which a first set of variables represents the positions and orientations of the magnetic objects P as well as the amplitudes of the magnetic moments of these objects.  A second set of variables represents the measurements of the magnetometers Mu.  To obtain the positions, orientations and amplitudes of the magnetic moments of the P magnetic objects, the variables of the first set are the unknowns and the values of the variables of the second set are known.  This model is typically constructed from the physical equations of magnetism.  This model is parameterized by the known distances between the magnetometers Mu.  Here, magnetic objects are permanent magnets.  To build this model, each permanent magnet is approximated by a magnetic dipole.  This approximation introduces very few errors if the distance between the permanent magnet and the magnetometer Mu is greater than 2L and preferably greater than 3L, where L is the largest dimension of the permanent magnet.  Typically, L is less than 20 cm and preferably less than 10 or 5 cm.  Here, the Mp model is nonlinear.  The unit 30 solves it by implementing an algorithm for estimating its solution.  For example, the algorithm used is an ensemble Kalman filter better known by the term "Unscented Kalman Filter".  Since each magnetic object is characterized by three variables to know its position, two variables to know its orientation and a variable to know the amplitude of its magnetic moment, the maximum number of magnetic objects simultaneously locatable by the Network of N magnetometers is less than N / 2.  Therefore, the value of the number P is less than or equal to N / 2 and preferably less than N / 5 or N / 10 or N / 20 to have redundant measurements.  The redundancy of the measurements makes it possible to improve the precision of the location of the magnetic objects.  In this embodiment, only five mathematical models M1, M2, M3, M4 and M5 are implemented in the unit 30 for, respectively, 1, 2, 3, 4 and 5 permanent magnets simultaneously present in front of the face 24. .  The unit 30 is also capable of transmitting a command to the device 15 4 via an interface 36 connected to this device 4.  The memory 34 also comprises a database 38 in which are recorded several control laws of the device 4.  Each control law makes it possible to generate the control of the apparatus 4 corresponding to the current state of the utensil with which it is associated.  For this purpose, each control law associates: - at least one possible state of the utensil, a control of the apparatus 4, and - to another possible state of the same utensil, another control of the apparatus 4 or lack of control of the device 4.  The control law therefore determines how the apparatus 4 operates in response to the actuation by the user of the utensil which is associated with this control law.  This base 38 is described in more detail with reference to FIG.  FIG. 2 shows part of the magnetometers Mu of the device 12.  These magnetometers Mu are aligned in lines i parallel to the direction X.  These magnetometers are also aligned in columns parallel to the Y direction to form a matrix.  The lines i and the columns j are arranged in the order of the increasing indices.  The center of the magnetometer Mu is at the intersection of line i and column j.  The center of the magnetometer corresponds to the point where the magnetic field is measured by this magnetometer.  Here, the indices i and j belong to the interval [1; 8].  The centers of two magnetometers Mu and Mu. , immediately consecutive along a line i are separated by a known distance d ,,,,,. 1.  Similarly, the center of two magnetometers Mu and M, + 1 ,,, immediately consecutive along the same column j are separated by a known distance dJ, ',. 1.  In the particular case described here, whatever the line i, the distance d ,,,, J + 1 is the same.  This distance is therefore noted d ,.  Similarly, whatever the column j, the distance dj, ',. 1 between two magnetometers is the same.  This distance is therefore noted d ,.  Here, the distances d, and dj are both equal to d.  Typically, the distance d is less, and preferably two times smaller, the smallest distance that can exist between two magnetic objects simultaneously present in front of the face 24 during normal use of the interface 2.  Here, the distance d is between 1 and 4 cm when: the power of the permanent magnet is 0.5 A. m2, - the sensitivity of the magnetometers is 4 * 10-7T, and - the number of magnetometers Mu is sixty-four.  Figures 3 and 4 show other utensils usable in place of the utensil 10 or simultaneously with the utensil 10 in the interface 2.  FIG. 3 represents a cursor 90.  This slider makes it possible, for example, to set the value of a parameter of the device 4.  This slider 90 comprises for this purpose a slider 92 mounted displaceable in translation in a rectilinear slit 94 formed in an upper face of a housing 96.  The user can manually slide the slider 92 to change the state of the slider 90.  In the case of the slider 90, each state corresponds to a particular position of the slider 92 along the slit 94.  To measure the position of the slider 92 along the slot 94, the slider 90 is equipped with: - a permanent magnet 98 fixed without any degree of freedom to the slider 92, and - a permanent magnet 100 fixed without no degree of freedom to the housing 96.  We define a local orthogonal reference frame Rgo bound without any degree of freedom to the housing 96 of which: - the axis Xgo is parallel to the slot 92, - the axis Ygo is parallel to the bottom of the housing 96, - Zgo axis is perpendicular to the bottom of the housing 96, and 30 - the origin is coincident with the geometric center of the magnet 100.  For the sake of clarity, the reference Rgo is shown next to the cursor 90 in FIG. 3.  In this embodiment, the direction of the magnetic moment of the magnet 98 is parallel to the direction x9o.  The magnetic moment direction of the magnet 100 is parallel to the z90 direction.  The number of degrees of freedom of the slider 92 relative to the housing 96 is limited to a degree of freedom in translation along the axis Xgo.  Under these conditions, the following additional equations exist between the coordinates of the magnets 98 and 100 in the frame Rgo Z98 = Z100 + h98 and y98 = y100, where: - x98, y98, z98 are the coordinates in the frame Rgo of the geometric center of the magnet 98, - xioo, yloo, zioo the magnet 100, - h98, is a constant numerical value known as a function of the dimensions of the utensil 90.  We also have the following additional equations: e-y98 = ey100 and ez100 = ez98 + 90 °, where: - ey98 and ez98 are the angles between the direction of the magnetic moment of the magnet 98 are the axes y90 and z90 and - ey100 and ez100 are the angles between the direction of the magnetic moment of the magnet 100 are the y90 and z90 axes.  The amplitudes of the magnetic moments of the magnets 98 and 100 are unique in the interface 2 and different from each other.  [0061] FIG. 4 represents a button 110.  This button 110 is manually movable by a user between a state of rest (shown in FIG. 4) and a depressed state.  Typically, the button 110 is intended to trigger an action of the device 4 only when it reaches its depressed state.  This button 110 comprises a key 112 only displaceable in translation, along an axis 114 secured to a housing 116, between a rest position (shown in Figure 4) and a depressed position.  In Figure 4, the axis 114 is vertical.  The idle and depressed states correspond, respectively, to the idle positions and pressed from the key 112.  The button 110 comprises a spring 118 interposed between a lower portion of the button 112 and an upper face of the housing 116.  A straight rod 120 extends along the axis 114, this rod 120 is fixed on one side, without any degree of freedom, to the key 112 and, on the other side, to a stop 124.  This rod 120 is slidably mounted inside an orifice 122 formed in the upper face of the housing 116.  The stop 124 makes it possible to retain the lower end of this rod 120 inside the casing 116.  To determine the state in which the button 110 is located, the rod 120 comprises a permanent magnet 126 fixed without any degree of freedom on this rod.  In this embodiment, the direction of the magnetic moment of the magnet 126 coincides with the axis 114.  Another permanent magnet 128 is also fixed without any degree of freedom on the bottom of the housing 116.  Therefore, the state in which the button 110 is located can be established from the relative position of the magnet 126 relative to the magnet 128.  For example, the state is established from the value of a distance d110 and a threshold So.  The distance d110 is the shortest distance between the geometric centers of the magnets 126 and 128.  are the coordinates in the Rgo coordinate system of the geometric center of the So threshold is used to discriminate the depressed state of the state of rest.  If the duo distance is less than the threshold So, this means that the button 110 is in its depressed state.  Conversely, if the distance d110 is greater than this threshold So, the button 110 is in its idle state.  We define a local orthogonal reference R110 fixed without any degree of freedom to the button 110.  For the sake of clarity, it is shown next to the button 110 in FIG.  The origin of the reference R110 is confused with the geometric center of the magnet 128.  The axes X110 and Z110 of this reference R110 coincide with the directions of the magnetic moments, respectively, of the magnets 128 and 126.  The axis Z110 of this reference R110 is thus coincident with the axis 114.  The magnet 126 has only one degree of freedom in translation relative to the magnet 128.  Under these conditions, the following additional equations interconnect the coordinates of the magnets 126 and 128 in the frame R110: x126 = X128, Y126 = Y128, eY126 = eY128, and ez126 = OZ128 + 90 °, where: - x126, Y126, 2126 are the coordinates in the reference R110 of the geometric center of the magnet 126, - X128, Y128, 2128 are the coordinates in the reference R110 of the geometric center of the magnet 128.  Here, the magnets 126 and 128 are identical.  Figure 5 shows the database 38.  This base 38 comprises for each utensil U: - a selection condition C, for selecting, from a distinctive characteristic of the utensil U 'only this utensil U, and not the other utensils having an arrangement of different magnets and capable of being used in the interface 2, - a control law L, prerecorded, - a set E, of additional equations defining the predetermined relationship (s) that exist between the coordinates and the orientations of the magnetic moments magnets of this utensil.  Subsequently, it is assumed that the distinctive characteristic of each utensil is the amplitude of the magnetic moment (s) of its permanent magnets.  [0071] Preferably, the selection condition is expressed in the form of a range of values to which the distinctive characteristic of the utensil must belong to be verified.  If the condition C, is verified by a permanent magnet or a pair of permanent magnets, then the law L, and the set E, are associated with the utensil equipped with this magnet or pair of magnets.  The index "i" identifies the law L, and the set E, associated with this condition C, by the base 38.  Here, each condition C relates to the values of the variables estimated for the permanent magnets which equip a utensil and which, generally, are not used to establish the state in which this utensil is currently located.  By way of illustration, an example of content of the base 38 will now be described.  The law Li is a control law that uses only the values of the variables ey and Oz.  More precisely, it transforms each value of the angles Oy and Oz into a respective command of the apparatus 4.  For example, this command can cause a displacement of the character 6 in a direction specified by the values of angles ey and Oz.  This law Li is associated with a condition Ci.  The condition Ci is as follows: the value of the variable A must be between Si and 52, where Si and 52 are predefined limits such that only the amplitude of the magnetic moment of the magnet 16 is between these limits.  This law Li is therefore intended to be associated with the utensil 10.  If the device 12 determines that the value of the variable A of a permanent magnet is between S1 and S2, then it uses the law Li to convert the values of the angles ey and Oz of the same magnet into a command the apparatus 4 .  Therefore, by changing the inclination of the utensil 10, it is possible to control the apparatus 4.  This remains true regardless of the position of the utensil 10 relative to the device 12 from the moment when the presence of the magnet 16 can be detected.  The location of the utensil 10 can therefore freely be chosen by the user, which corresponds to an infinity of possible configuration for the interface 2.  The set E1 comprises the additional equation described with reference to FIG.  The law L2 is a control law which uses the value of a distance between two permanent magnets to generate a setting control of a parameter 25 of the apparatus 4 as a function of this distance.  This law L2 specifies that the distance is measured between: a permanent magnet whose amplitude of the magnetic moment is between predefined limits S71 and S81 such that only the magnet 98 can satisfy this condition, and 30 - a permanent magnet whose the magnitude of the magnetic moment is between predefined limits S72 and S82 such that only the magnet 100 can satisfy this condition.  This control law is typically designed to be associated with the cursor 90 so that the distance corresponds to the distance d90.  For this purpose, the law L2 35 is associated with a condition C2.  This condition C2 is as follows: the value of the variable A of a permanent magnet is between the limits S72 and S82.  With such a condition C2, the law L2 is associated with the cursor 90.  The set E2 comprises the additional equations described with reference to FIG.  The law L3 is a control law that uses only the shortest distance between two identical permanent magnets to generate a control of the apparatus 4.  Here, the law L3 is intended to be associated with the button 110 so that the shortest distance corresponds to the distance d110.  For example, the control law L3 is as follows: if the distance d110 is greater than the threshold So, then the unit 30 generates no control of the apparatus 4, and if the distance d110 is less than or equal to the threshold So, then the unit 30 generates and transmits a command to the apparatus 4.  The law L3 is associated with a condition C3.  Here, this condition C3 is as follows: the values of the variables A of two of distinct permanent magnets are both between limits S9 and Sl0, and the shortest distance between these two magnets is less than a threshold D max110.  The limits S9 and Slo are predefined constants such that only the amplitude of the magnetic moment of the magnets 126 and 128 is between these limits.  The value of the threshold Dmax110 is chosen equal to the largest possible value of the distance d110.  The set E3 comprises the additional equations described with reference to FIG.  The operation of the interface 2 will now be described in more detail with reference to the method of FIG. 6.  This process starts with an initialization phase 140 in which the laws L 'the sets E, and the conditions C, are recorded in the base 38.  Then, we proceed to a phase 142 of configuration of the interface 2.  During this phase 142, the user chooses one or more utensils from utensils 10, 90 and 110.  Here, it is assumed that it has only one copy of the utensil 10 and two copies of the utensils 90 and 110.  Then, he disposes them freely on the face 24 of the device 12.  In addition, he ensures that the configuration made does not lead to simultaneously having more than five permanent magnets in front of the face 24.  The human-machine interface 2 is then configured.  The following description is made in the particular case where the user has disposed on the face 24 a copy of each of utensils 10, 90 and 110.  It is then possible to use this interface 2 to control the device 4.  This begins with a step 146 in which the magnetometers Mu simultaneously measure the magnetic field of the magnet or magnets simultaneously present in front of the face 24.  Then, in a step 148, the unit 30 estimates from the magnetometer measurements Mu the position, the orientation and the amplitude of the magnetic moment of each of the magnets present.  For this, during an operation 150, the unit 30 solves the system of equations 5 of the model M1 to a magnetic dipole.  It obtains a set of coordinates x ,, yi, z1, 01 and cp, and an amplitude Al corresponding to the estimate of the position, the orientation and the amplitude of a single permanent magnet.  Then, during an operation 152, the unit 30 calculates an estimation error E1 representative of the difference between: 10 - estimated values of the magnetometer measurements, calculated from the system of equations M1 and from the position, orientation and amplitude obtained at the end of the operation 150, and the measurements of the magnetometers recorded during the step 146.  The error is estimated for each magnetometer.  Here, the overall estimation error resulting from the use of the M1 model is for example obtained by averaging the calculated errors for each of the magnetometers.  In the case where the algorithm used to solve the model is an ensemble or extended Kalman filter, the resolution of this system of equations during the operation 150 also provides an estimate of this error El.  Steps 150 and 152 are performed for P = 1 to P = 5.  Preferably, steps 150 and 152 for each value of P are performed in parallel.  Then, during an operation 154, the unit 30 selects the model Mp which gives the error Ep the smallest.  Thus, if there is only one permanent magnet in front of the face 24, the unit 30 automatically selects the model M1.  If, on the contrary, there are two permanent magnets, the unit 30 automatically selects the model M2 and so on.  Here, the M5 model is automatically selected because the interface 2 has five permanent magnets.  This M5 model has thirty variables whose values are unknown.  At the end of this step 148, the computer has a first estimate 30 of the values of the variables x, y, z, ey, 02 and A for each of five magnets.  This first estimate is that obtained following the execution of operation 150 using the model M5.  Step 148 then continues with a first phase 166 of automatic adaptation of the model M5 to the configuration performed by the user during step 142.  This phase 166 begins with a step 168 of identification of sets E, which can be used to reduce the number of variables to be estimated in the M5 model.  For this purpose, during step 168, the computer 32 identifies the utensils present in front of the face 24.  For example, during an operation 170, the computer acquires the distinctive characteristic of each utensil present in front of the face 24.  Here, this distinctive characteristic is obtained from the previous estimate of the position, orientation and amplitude of each permanent magnet.  In this particular case, this distinctive characteristic is either the amplitude of the estimated magnetic moment of a permanent magnet, or the estimated magnitudes of the magnetic moments of a pair of permanent magnets.  The estimated amplitudes are for example those estimated during the execution of the operation 150 with the model M5.  Then, during an operation 172, the calculator 32 compares each distinctive characteristic acquired with each condition C, of the base 38.  If the distinctive characteristic acquired corresponds to one of the conditions C, then, during an operation 176, a utensil is identified.  Each time a cookware is identified, an identifier of that cookware is added to a list of identified cookware.  This list of identified utensils also contains, for each utensil 15 identified, the set E, and the law L, associated with this utensil by the base 38.  [00100] If the distinctive characteristic acquired does not correspond to any of the conditions C 'then the utensil is not identified.  At the end of step 168, the additional equations that can be used to reduce the number of variables of the M5 model are therefore identified.  However, each of these additional equations, expressed in the frame Rm, depends on the coordinates of the origin and the orientation of the axes of the local coordinate system relative to the reference Rm.  Therefore, the method continues with a step 178 in which, for the local reference of each identified utensil, the calculator 32 determines: 25 - the position of its origin in the reference Rm, and if necessary, - orientation of its axes in the reference Rm.  For this, the computer 32 uses the position and the estimated orientation of the permanent magnets of the utensil in the reference Rm and prerecorded information on the position of the local coordinate system relative to the permanent magnets 30 of this utensil.  For example, in the case of the utensil 10, the position of the origin of the reference R10 is determined by translating the estimated position of the magnet 16 in the reference Rm of a distance Rd in the direction of its magnetic moment and approaching the face 24.  In the case of the utensil 90, the position of the origin of the reference Rgo in the frame Rm is equal to the estimated position of the geometric center of the permanent magnet 100.  The orientations of the Zgo and Xgo axes are equal to the estimated orientations of the magnetic moments, respectively, of the magnets 100 and 98.  Similarly, the computer 32 determines the position and the orientation of the local coordinate system R11o.  At the end of step 178, each of the additional equations identified expresses equality between a first variable of the M5 model and a term.  The value of this term is expressed using one or more other variables of the M5 model different from the first variable.  Typically, these other variables of the M5 model are those that encode the position and orientation of the permanent magnets of the same utensil.  Then, for each additional equation, during a step 180, the calculator 32 replaces, in the current system of equations of the M5 model, the first variable by the term to which it is equal.  Thus, at the end of step 180 we obtain a new system of equations and therefore a new current model, noted M5c, with fewer variables to estimate than the current system.  For the steps that follow the system of equations of the Mc model becomes the current system of equations.  More precisely, the initial M5 model is a generic model that makes it possible to simultaneously locate five permanent magnets without taking into account the mechanical relationships that may exist between these permanent magnets.  This model M5 is therefore a system of equations that has thirty variables to estimate.  After the execution of step 180, the new model Ms, is a specific model which depends on the utensils currently arranged on the face 24.  This model M5, in particular takes into account the mechanical relationships that limit the number of degrees of freedom moving the permanent magnets relative to each other.  It therefore has fewer variables to estimate than the M5 model.  For example, here the Ms model has only 19 variables to estimate instead of thirty.  Indeed, even before having determined whether utensils 10, 90 and 110 are immobile or not in the frame Rm and to have replaced their magnetic moment by constants, it is possible to simplify the current system of equations using the three additional equations of the utensil 10 and the four additional equations of each of the utensils 90 and 110.  Subsequently, the replaced variables are directly calculated from the additional equations identified and the estimated values for the variables of the M5c model.  The process continues with a step 190 in which the magnetometers 35 Mu measure the magnetic field present.  In a step 192, the computer 32 acquires these measurements.  In a step 194, the calculator 32 estimates the position and the orientation of each permanent magnet by solving the system of equations of the M5c model.  This system of equations is for example solved as previously described during the operation 150.  In a step 196, the calculator 32 also calculates the current estimation error Eerr related to the use of the M5c model.  This error Eerr is calculated as described in step 150 except that the Msc model is used.  Then, during a step 198, the error Eerr is compared to a predetermined threshold Serr.  If the error Eerr is below this threshold Serr, then the computer 32 proceeds to a step 200.  In step 200, the laws L, selected and the positions and orientations 10 estimated during step 194 are used to generate one or more commands transmitted to the device 4.  Then, after step 200, the process returns to step 190.  [00117] If the error Eerr is greater than the threshold Serr, then a step 202 of reintroduction of one or more variables in the current model Msc is carried out to reduce this error Eerr.  Step 202 can be performed in many different ways.  Indeed, there are many methods to reintroduce into the current model one or more variables so as to limit the error Eerr.  For example, a first method is to always return to step 146.  In this case, all the steps from step 146 are reiterated each time the error Eerr exceeds the threshold Serr.  This method has the advantage of allowing to reintroduce all the variables of the models M1 to M5.  Moreover, it is simple to implement.  [00119] A second method consists of reintroducing the variables one after the other in the reverse order where they have been eliminated from the current model.  Thus, the computer 32 reintroduces into the current system of equations the last variable eliminated so as to obtain a new system of equations containing one more variable.  Then, this new system of equations is used during the following implementations of steps 192 to 198.  This second method has the advantage that it is not necessary to systematically repeat step 148 and phase 166 of automatic adaptation of the model.  Another advantage of this method is that the last fixed variable is the most likely to be modified.  A third method consists in testing the reintroduction in the current model of each variable and only reintroducing the variable or the variables which make it possible to reduce the error Eerr.  For example, for this, the computer 32 first reintroduces a first variable into the current system of equations, then checks whether the error Eerr calculated with the new system of equations thus obtained is lower than that obtained during the last iteration of step 196.  If the error decreases, then this variable is selected.  On the other hand, if the error is not modified or increases, then this variable is not selected.  The computer 32 thus tests each of the variables of the current model Ms, one after the other.  Then, it reintroduces into the current model Ms, all the selected variables.  If the implementation of the second or the third method leads to reintroduce into the current model all the variables and that despite this, the error Eerr remains above the threshold Serr, then the process returns to step 148.  Indeed, it may mean that the increase in error Eerr may then be caused by the removal or addition of a new utensil on the face 24 which changes the number of permanent magnets present.  In parallel with the steps 196 to 200, during a step 206, the user manipulates the utensils 10, 90 and 110 to control the apparatus 4.  Also in parallel with steps 196 to 200, the computer 32 executes a second phase 210 of automatic adaptation of the current model.  This phase 210 begins with a step 211 of identifying an additional equation.  In this case, in step 211, the additional equation is constructed and not obtained from prerecorded information on the moving objects.  For this purpose, during an operation 212, the calculator 32 calculates the amplitude of the variations of each variable of the model Ms, during the preceding Q iterations of the steps 190 to 194.  Typically Q is an integer greater than ten and preferably greater than one hundred or one thousand.  For example, during the operation 212, the calculator 32 calculates the variance of each of the variables of the model M5c.  Then, during an operation 214, the computer 32 compares each calculated variance with a respective predetermined threshold Svar.  For example, the threshold Svar is equal to a fraction of the maximum variance measured in the past for this variable.  For example, this threshold Svar is taken equal to 10% or 5% or 1% of this maximum measured variance.  The threshold Svar can also be a predetermined numerical constant.  If the variance calculated during the operation 212 is greater than this threshold Svar, then it is considered that the value of this variable varies and no new additional equation is identified.  In the contrary case, the calculator 32 considers that this variable does not vary any more, which makes it possible to establish a new additional equation.  This additional equation is an equality relation between this variable and a numerical constant.  The value of this numerical constant is for example taken equal to the average of several of the previous values estimated for this variable or simply taken equal to the last estimated value for this variable.  Then, during a step 216, the variables that are considered to be constants are replaced in the Ms model by the values of these constants.  We then obtain a new system of equations with even fewer variables than the current system of equations.  This new system of equations is then used during the following executions of steps 190 to 198.  It replaces the previous system of equations.  Thus, the phase 210 allows to adapt the current model by exploiting the fact that in practice, a degree of freedom of a utensil is not necessarily used by the user.  The additional equations thus constructed are added to the list of additional utensils and equations already identified.  These variables eliminated during the phase 210 can be reintroduced into the system of equations during the step 202.  [00130] When the user wishes to reconfigure the interface 2.  He returns to phase 142.  During this new execution of the phase 142, it can modify the configuration of the interface 2, for example: - by removing or adding a utensil, and / or - by moving a utensil.  [00131] Thus, the interface 2 is easily configurable while remaining simple to manufacture.  [00132] Many other embodiments are possible.  For example, the front face 24 is not necessarily flat.  For example, alternatively, it is a relief shape.  Typically, this relief shape may include recessed housing for receiving and immobilizing the utensils in these dwellings.  [00133] Other utensils than those described above are feasible and usable in the interface 2.  For example, reference may be made to this subject to the patent application FR1354160 filed May 7, 2013 by the Commissariat à l'énergie atomique et aux energies alternatives.  Many other distinguishing features and associated selection conditions can be devised.  For example, when the implement has K permanent magnets, with K strictly greater than one, the distinctive characteristic (s) of the utensil are selected from the group consisting of: - the relative position of the K permanent magnets with respect to other, - the orientation of the magnetic moments of the K permanent magnets with respect to each other, - the magnitudes of the magnetic moments of the K permanent magnets or the relative magnitudes of the magnetic moments of the K permanent magnets relative to each other.  When the distinguishing feature is the position of a permanent magnet of a utensil relative to the position of another permanent magnet of the same utensil, it is not necessary to use the amplitude of the magnetic moments for select the set Ei to associate with this utensil.  When the utensil comprises only a permanent magnet, the distinctive characteristic of the utensil is the amplitude of the magnetic moment of its permanent magnet.  [00137] The distinctive characteristic may also be the position of a permanent magnet in the reference Rm.  In this case, the selection condition checks whether this permanent magnet is located inside a predefined zone in the reference Rm.  If so, a control law and a set of additional equations are associated with this utensil.  It is also possible to define conditions verifying whether the position of a magnet is not located within a predefined volume but in a particular position with respect to a fixed point of the mark Rm or at a point whose position is defined by another permanent magnet distinct from the utensil.  The distinguishing characteristic may also be the inclination of the magnetic moment of a permanent magnet with respect to a reference axis of the reference Rm, for example the vertical.  In this case, the selection condition tests whether the estimated tilt of a magnetic moment of a permanent magnet of the utensil is within a predetermined range of values.  If so, the calculator identifies the set E of additional equations associated with this selection condition as to reduce the number of variables in the current system of equations.  The acquisition of a distinctive feature of an object can be achieved differently from what has been described.  For example, each object is equipped with an RFID tag (Radio Frequency Identification).  In this case, the device 12 further comprises an RFID tag reader.  Thus, when a utensil is presented in front of the device 12, it is identified from the reading of the identifier of this utensil contained in its RFID tag.  Then, this identifier is used to select in a prerecorded database the set E, associated with this utensil.  This same identifier can also be used to select the control law or laws to be used with this utensil.  Thus, the distinguishing feature is not necessarily a feature constructed from the Mu magnetometer measurements.  The acquisition of the distinctive characteristic is not necessarily automatic.  For example, alternatively, the user indicates, via a man-machine interface, the identifier of the object or objects added or removed.  In one embodiment where the interface comprises only one or more identical copies of the same utensil, the selection conditions may be omitted.  Indeed, in this case, the set E of additional equations associated with each utensil is always the same.  The database 38 may then contain only one record.  Simply count the number of permanent magnets.  The database 38 does not necessarily contain, in addition to sets E ', the control laws L.  For example, base 38 contains only conditions C, and sets E, and another database contains conditions C ', and laws L.  In this case the conditions C and C 'are not necessarily the same.  Thus, a set E is not necessarily and systematically associated with the same control law.  If the number P of permanent magnets simultaneously present in front of face 24 is known in advance, the control method can be simplified by using only the corresponding Mp model to determine position, orientation and amplitude. the magnetic moment of these permanent magnets P.  For example, the number P of permanent magnets is entered by the user during the configuration phase.  Step 148 can therefore be simplified.  The configuration change of the interface 2 is not necessarily performed directly by hand by a user.  In another variant, the interface comprises electric actuators for modifying the configuration of the interface.  For example, for this purpose, the movement of utensils or the replacement of these utensils by other utensils is motorized, so that the user does not himself have to handle each of these utensils.  It is possible to limit the number of locations where a utensil can be arranged relative to the front face 24.  For this, the interface 2 has several locations, each equipped with a key to prevent the attachment to this location of a number of utensils and, on the contrary, to allow the attachment to this location other different utensils.  The approximation used to construct the Mp model can also be a quaternary or higher approximation, that is to say that the equations of magnetism are approximated to an order greater than that corresponding to the dipolar approximation.  Many different methods can be used to determine the position and orientation of the magnetic object.  For example, the method described in US6269324 is usable.  These methods do not necessarily use a Kalman filter.  For example, the methods described in US2002 / 171427A1 or US6263230B1 are possible.  The magnetometers of the magnetometers network are not necessarily arranged in columns and in rows.  They can be arranged also according to other reasons.  For example, the magnetometers are arranged on each vertex of each triangular or hexagonal mesh of a mesh of a plane.  The arrangement of the magnetometers with respect to each other may also be random or non-regular.  Thus, the distance between two immediately consecutive magnetometers in the array is not necessarily the same for all pairs of two immediately consecutive magnetometers.  For example, the density of magnetometers in a given area of the network may be greater than elsewhere.  Increasing the density in a given area may increase the accuracy of the measurement in that area.  [00150] The magnetometers network can also extend in three non-collinear directions of space.  Under these conditions, the magnetometers are distributed within a three-dimensional volume.  The number N of magnetometers may also be greater than or equal to sixty-four or ninety.  [00152] Not all magnetometers in the magnetometer array are necessarily identical to each other.  Alternatively, the magnetometers do not all have the same sensitivity.  The device 4 can be replaced by any type of electrical device that must be controlled in response to an action of a human being.  For example, the controlled device may be a robot, a machine tool or the like.  In all of the embodiments described herein, the permanent magnet can be replaced by a permanently unpowered magnetic object that acquires a magnetic moment present in an external continuous magnetic field, such as the Earth's magnetic field. .  For example, the permanent magnet is replaced by a piece of soft magnetic material.  A magnetic material is considered soft if its coercive magnetic field is less than 10 or 1 A. m-1.  Such a piece 25 has a magnetic moment created by the interaction between the earth's magnetic field and the piece of soft magnetic material.  [00155] Many other embodiments of the method of FIG. 6 are possible.  For example, either of the phases 166 and 210 is omitted.  The number of measurements used to calculate the variance during the operation 212 may be a function of the duration of a sliding time window.  Typically, only the measurements made within this sliding time window are taken into account to calculate the variance.
    权利要求:
    Claims (10)
    [0001]
    REVENDICATIONS1. A method of locating mobile magnetic objects presented in front of a magnetometers network comprising N tri-axis magnetometers (Mu) mechanically connected to each other without any degree of freedom to maintain a known distance between each of these magnetometers, where N is a number an integer greater than or equal to five, the method comprising: a) measuring (190) by each of the magnetometers of the magnitude of the magnetic field along each of its measurement axes, b) the estimate (194) from the measurements of the magnetometers, values of several variables by solving a current system of equations connecting these variables to each measurement of a tri-axis magnetometer of the network, each variable corresponding to the position or the orientation or the amplitude of the a magnetic moment of a magnetic object, the iteration of steps a) and b) at successive instants in time to obtain the estimated values of each variable at these diffs ent successive instants, characterized in that the method includes: c) identifying (168; 211), according to the moving objects presented in front of the magnetometers network, of an additional equation connecting, by a relation of equality, a first variable of the current system of equations to a term, this term being a predefined relation between a or several other variables of the current system of equations or a constant numerical value, d) replacing (180; 216), in the current system of equations, of this first variable by the term to which it is equal, identified during the step c), to obtain a new system of equations in which the number of variables whose value is to be estimated is smaller than in the current system of equations, e) the use of the new system of equations to the place of the current system of equations during the following iterations of steps a) and b), the new system of equations thus becoming, for the following iterations of steps a) and b), the current equation system. ns.
    [0002]
    2. The method of claim 1, wherein the identification (168; 211) of the additional equation is performed based on the estimated values, during a previous iteration of steps a) and b) variables. 35
    [0003]
    3. Method according to claim 2, wherein the identification of the additional equation comprises: calculating (212) a magnitude representative of the amplitude of the variations of the first variable during previous iterations of the steps a) and b) from the 5 values of this first variable estimated during these previous iterations of steps a) and b), - comparing (214) this calculated quantity with a predetermined threshold to establish if this first variable has varied at during these previous iterations of steps a) and b), and 10 - if this predetermined threshold is crossed, the establishment (214) of a new additional equation connecting, by a relation of equality, the value of this first variable at a constant value, the constant value being a function of at least one value of this first variable estimated during these previous iterations of steps a) and b) during which it has been established that its value does not vary, then the use of this new additional equation in step d), and - if this predetermined threshold is not crossed, the lack of establishment of this new additional equation.
    [0004]
    4. A method according to any one of the preceding claims, wherein step c) comprises: acquiring (170) a distinctive feature of the moving magnetic object currently presented in front of the magnetometer array, and the comparison (172) of the distinctive characteristic acquired with pre-recorded selection conditions in a database, this database associating with each selection condition at least one additional equation connecting, by an equality relation, a variable of the current system of equations to a term, which term is a predefined relationship between one or more other variables of the current system of equations or a constant numerical value, and - only if the distinguishing characteristic acquired corresponds to one of the 30 conditions of pre-recorded selection, then the use in step d) of the additional equation associated with this selection condition. we.
    [0005]
    The method of claim 4, wherein acquiring a distinguishing feature comprises obtaining (170) this distinguishing feature from the 35 values of the estimated variables using the current system of equations.
    [0006]
    6. Method according to any one of the preceding claims, wherein the method comprises: calculating (198) an estimation error with the current system of equations, the estimation error being representative of the difference between: - estimated values of the magnetometer measurements when the position, orientation and amplitude of the magnetic moment of each magnetic object are equal to those estimated using the current system of equations during an iteration of the steps a ) and b), and - the measurements of the magnetometers recorded in step a) during the same iteration of steps a) and b), and - if the estimation error calculated exceeds a predetermined threshold: - the selection ( 202) of at least one of the additional equations previously used in step d) to obtain the current system of equations, - replacing (202), in the current system of equations, with the term of the additional equation selected, p ar the variable that this term has replaced to obtain a new system of equations in which the number of variables whose value is to be estimated is greater than in the current system of equations and the use of the new system of equations to the place of the current system of equations during the following iterations of steps a) and b), and if the calculated error does not cross this predetermined threshold, maintaining the use of the current system of equations during the following iterations steps a) and b).
    [0007]
    7. The method of claim 6, wherein the selection of at least one additional equation comprises: the selection of at least two additional equations previously used in step d) to obtain the current system of equations, for each of the selected additional equations, the replacement in the current system of equations of the term of the selected additional equation by the variable that this term has replaced to obtain third is four different systems of equations, each of these fourth and third systems of equations having a number of variables whose value must be estimated greater than in the current system of equations, - the calculation of the estimation error with these third and fourth systems of equations, and the selection of the additional equation, among those initially selected, which minimizes the calculated estimation error.
    [0008]
    8. Information recording medium (34), characterized in that it comprises instructions for the execution of a method according to any one of the preceding claims, when these instructions are executed by an electronic calculator .
    [0009]
    9. Device for locating mobile magnetic objects, this device comprising:
    [0010]
    An array of magnetometers having N tri-axis magnetometers (M, J) mechanically connected to each other without any degree of freedom to maintain a known distance between each of these magnetometers, where N is an integer greater than or equal to five; a processing unit (30) capable of: a) acquiring the measurements of each of the magnetometers of the magnetometer array; b) estimating, from the magnetometer measurements, the values of several variables by solving a current system of equations connecting these variables to each measurement of a tri-axis magnetometer of the array, each variable corresponding to the position or orientation or magnitude of a magnetic moment of a magnetic object, - to repeat the steps a) and b) at successive instants in time to obtain the estimated values of each variable at these different successive instants, characterized in that the processing unit (30) is also able: c) to identify, according to the moving objects presented in front of the magnetometers network, an additional equation connecting, by a relation of equality, a first variable of the current system of equations to a term, this term being a relation predefined between one or more other variables of the current system of equations or a constant numerical value, 30 d) to replace, in the current system of equations, this first variable by the term to which it is equal, identified during the step c), to obtain a new system of equations in which the number of variables whose value is to be estimated is smaller than in the current system of equations, e) to use the new system of equations in place of the system of equations 35 during the following iterations of steps a) and b), the new system of equations thus becoming, for the following iterations of steps a) and b), the current system of equations ons.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP2884372B1|2017-02-01|Method for locating mobile magnetic objects presented before an array of magnetometers
    EP2923227B1|2020-02-12|Method for automatically recognising a moving magnetic object
    WO2013144342A2|2013-10-03|Method and device for locating a magnetic object
    FR2980327A1|2013-03-22|DEVICE AND METHOD FOR COLLECTING INFORMATION RELATING TO ACCESS POINTS
    EP2779091B1|2015-08-19|Automatic stereoscopic camera calibration
    FR3013488A1|2015-05-22|METHOD OF ESTIMATING THE SPEED OF MOVING A CAMERA
    EP2965106B1|2019-08-14|Method for automatically recognising a magnetic object
    EP2831705B1|2020-07-22|Screen with magnetic object locating
    FR3054040A1|2018-01-19|METHOD FOR ESTIMATING AN AVERAGE ANGULAR DEVIATION BETWEEN THE MAGNETIC AXIS AND A REFERENCE AXIS OF A MAGNETIC OBJECT
    WO2015071457A1|2015-05-21|Method for estimating the speed of movement of a camera
    EP3360031B1|2019-09-04|Method of tracking a position of a magnet by differential measurement
    FR3072180B1|2019-11-01|METHOD FOR LOCATING A TERMINAL OF ACCESS TO A COMMUNICATION NETWORK
    FR3005516A1|2014-11-14|CONFIGURABLE MAN-MACHINE INTERFACE
    EP2889648B1|2016-08-03|Method for locating mobile utensils presented before an array of magnetometers
    FR2971859A1|2012-08-24|METHOD AND DEVICE FOR SMOOTHING WITH PRESERVATION OF THE TRAVEL TIME
    EP3230772B1|2022-02-23|Method for locating at least one movable magnetic object and associated system
    FR3067138B1|2019-07-26|METHOD FOR ESTIMATING THE POSITION OF A MAGNET COMPRISING AN IDENTIFICATION PHASE OF A MAGNETIC DISTURBATOR
    EP2994813B1|2019-02-20|Method for controlling a graphical interface for displaying images of a three-dimensional object
    FR3069068B1|2019-08-23|METHOD FOR LOCATING A EVOLVING OBJECT IN A MAGNETIC FIELD GENERATED BY A SET OF AT LEAST THREE MAGNETIC GENERATORS
    EP2791687B1|2017-06-21|Method and system for auto-calibration of a device for measuring speed of a vehicle travelling in a three-dimensional space
    FR2963422A1|2012-02-03|Method for estimating movement of person in building of city for indoor location of person, involves obtaining series of angular speed measurements and calculating frequency of steps completed by person from measurements
    Sung2015|Automated Reconstruction of 3D Object on Embedded System for Mobile Devices
    EP2587218A1|2013-05-01|Method for locating a device in an environment and corresponding device
    FR3075399A1|2019-06-21|METHOD FOR LOCATING A CONNECTED OBJECT CAPABLE OF TRANSMITTING AN ELECTROMAGNETIC SIGNAL BY A MOBILE TERMINAL FOR RECEIVING THIS ELECTROMAGNETIC SIGNAL
    EP3227743A1|2017-10-11|Method for calibrating an augmented reality visual rendering system comprising at least one display device that is partially transparent with respect to the user thereof, and associated system
    同族专利:
    公开号 | 公开日
    EP2884372A1|2015-06-17|
    US10436567B2|2019-10-08|
    FR3015049B1|2015-12-25|
    US20150168123A1|2015-06-18|
    JP2015122071A|2015-07-02|
    JP6424082B2|2018-11-14|
    EP2884372B1|2017-02-01|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    GB2310930A|1996-03-05|1997-09-10|Hughes Aircraft Co|Dipole moment detector and localizer|
    US20020171427A1|2001-04-24|2002-11-21|The United States Of America Represented By The Secretary Of The Navy|Magnetic anomaly sensing system and methods for maneuverable sensing platforms|
    US20030095115A1|2001-11-22|2003-05-22|Taylor Brian|Stylus input device utilizing a permanent magnet|
    US7932718B1|2009-03-12|2011-04-26|The United States Of America As Represented By The Secretary Of The Navy|System and method using magnetic anomaly field magnitudes for detection, localization, classification and tracking of magnetic objects|
    FR2952450A1|2009-11-12|2011-05-13|Commissariat Energie Atomique|PORTABLE DEVICE AND METHOD FOR PRINTING AN IMAGE, RECORDING MEDIUM, PEN AND TERMINAL FOR THIS DEVICE|
    FR2988862A1|2012-03-29|2013-10-04|Commissariat Energie Atomique|METHOD AND DEVICE FOR LOCATING A MAGNETIC OBJECT|WO2018011492A1|2016-07-13|2018-01-18|ISKn|Method for estimating an angular deviation between the magnetic axis and a reference axis of a magnetic object|FR1354160A|1963-01-11|1964-03-06|Air Liquide|Process for preparing oxidizing mixtures based on monoperphthalic acid|
    JP2751544B2|1990-03-27|1998-05-18|株式会社島津製作所|Magnetic detector|
    US6263230B1|1997-05-08|2001-07-17|Lucent Medical Systems, Inc.|System and method to determine the location and orientation of an indwelling medical device|
    US6269324B1|1998-10-19|2001-07-31|Raytheon Company|Magnetic object tracking based on direct observation of magnetic sensor measurements|
    FR2988874B1|2012-03-29|2014-03-28|Commissariat Energie Atomique|METHOD FOR DETECTING A POINT OF CONTACT BETWEEN A TIP OF A UTENSIL AND A WRITING MEDIUM|
    EP2645204B1|2012-03-29|2018-12-19|Deutsche Telekom AG|Accessory cover for an electronic device and system|
    FR2988872B1|2012-03-29|2014-03-28|Commissariat Energie Atomique|SCREEN WITH MAGNETIC OBJECT LOCATION|WO2016178257A1|2015-05-05|2016-11-10|Garofalo Enrico|Contactless detection system of the position of objects on a surface|
    FR3042290B1|2015-10-09|2018-10-12|ISKn|METHOD FOR TRACKING A POSITION OF A MAGNET BY DIFFERENTIAL MEASUREMENT|
    FR3067139B1|2017-05-31|2019-09-27|ISKn|METHOD FOR TRACKING A MAGNET BY A MAGNETOMETER NETWORK, COMPRISING A MAGNETIC DISPENSER IDENTIFICATION PHASE AND A MAGNETIC DISTURBATOR|
    FR3067138B1|2017-05-31|2019-07-26|ISKn|METHOD FOR ESTIMATING THE POSITION OF A MAGNET COMPRISING AN IDENTIFICATION PHASE OF A MAGNETIC DISTURBATOR|
    FR3080186B1|2018-04-16|2020-05-08|Commissariat A L'energie Atomique Et Aux Energies Alternatives|CALIBRATION METHOD OF A NETWORK OF MAGNETOMETERS|
    法律状态:
    2015-12-31| PLFP| Fee payment|Year of fee payment: 3 |
    2016-12-29| PLFP| Fee payment|Year of fee payment: 4 |
    2018-09-28| ST| Notification of lapse|Effective date: 20180831 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1362730A|FR3015049B1|2013-12-16|2013-12-16|METHOD FOR LOCATING MOBILE MAGNETIC OBJECTS PRESENTED BEFORE A MAGNETOMETER NETWORK|FR1362730A| FR3015049B1|2013-12-16|2013-12-16|METHOD FOR LOCATING MOBILE MAGNETIC OBJECTS PRESENTED BEFORE A MAGNETOMETER NETWORK|
    JP2014252117A| JP6424082B2|2013-12-16|2014-12-12|A method to detect moving magnetic objects in front of a network of magnetometers|
    EP14197663.9A| EP2884372B1|2013-12-16|2014-12-12|Method for locating mobile magnetic objects presented before an array of magnetometers|
    US14/571,726| US10436567B2|2013-12-16|2014-12-16|Method for locating mobile magnetic objects presented before a network of magnetometers|
    [返回顶部]