JP2006113019A - Triaxial type electronic compass, and azimuth detecting method using same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a triaxial type electronic compass capable of detecting precisely an azimuth in response to all the attitudes without being restricted by attitude angles, and an azimuth detecting method using the same. <P>SOLUTION: When two candidates α1, α2 exist in solution of the attitude angles calculated from magnetic data X, Y, Z, the azimuth is detected precisely in response to all the attitudes, by selecting the one proper attitude angle (α1 or α2) from on the first processing of selecting the proper attitude angle, based on a variation in the attitude angle calculated time-serially, the second processing of selecting the proper attitude angle, referring to the azimuth calculated just before and the attitude angle used in that time, and the third processing of selecting the proper attitude angle, on a ratio of the azimuth calculated just before to a differential value of the attitude angle used in that time. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a three-axis electronic compass for detecting an azimuth angle from geomagnetism and an azimuth detection method using the same, and more particularly to a three-axis electronic compass that can detect an azimuth angle without using a tilt sensor. The present invention relates to the direction detection method used.

  Prior art describing a method for detecting an azimuth angle in consideration of an error caused by a posture angle (posture angle with respect to the ground plane) when the compass itself is set in a tilted posture with respect to the ground plane For example, Patent Documents 1 and 2 shown below exist.

  In the device described in Patent Document 1, a magnetic sensor 100 that detects a biaxial component of a magnetic vector defined on a plane parallel to the substrate, and a Hall element 24 that detects a component in a direction perpendicular to the substrate of the magnetic vector. And a tilt sensor 22 that detects the posture angle of the substrate are integrally configured, and is detected from the magnetic sensor 100 based on the posture angle (roll angle or pitch angle) of the substrate detected by the tilt sensor 22. By correcting the biaxial component of the magnetic vector and the vertical component detected from the Hall element 24, an appropriate azimuth angle is detected.

In addition, in the earth magnetic pole direction calculation method for a magnetic sensor device described in Patent Document 2, a geomagnetic vector obtained from a three-axis magnetic sensor and a vector G having a predetermined angle α specified in advance are substantially orthogonal to a specific axis X. The predetermined angle α is approximately equal to the angle formed by the geomagnetic vector and the vertical vector from the ground, and the specific axis X is actually used by a board with a 3-axis magnetic sensor attached. In this situation, it is assumed that an axis substantially parallel to the ground among the axes used for main rotation is subjected to optimum inclination correction under use conditions that are frequently assumed.
JP 2002-196055 A JP 2003-166825 A

  However, the device described in Patent Document 1 requires a tilt sensor for performing tilt correction in addition to the three-axis magnetic sensor for calculating the earth magnetic pole direction. For this reason, the compass itself becomes a factor that increases the scale and weight, and as a result, there is a problem that the size and weight of a portable device or the like equipped with the compass is hindered.

  Moreover, in the thing described in the said patent document 2, since the fixed restriction | limiting (0-90 degree | times) was provided in the attitude angle, for example, the state where the user lay down on his back (therefore, the compass was mounted) When the mobile phone is used upside down, the correct azimuth angle cannot be obtained.

  The present invention is for solving the above-described conventional problems, and a triaxial electronic compass suitable for miniaturization and weight reduction without using a tilt sensor when acquiring an azimuth angle, and an azimuth detection method using the same. It is intended to provide.

  Another object of the present invention is to provide a three-axis electronic compass capable of detecting an azimuth angle with high accuracy corresponding to any posture without being limited by the posture angle, and an azimuth detection method using the same. It is said.

The present invention includes an (x′y′z ′) orthogonal coordinate system including three axes (x ′ axis, y ′ axis, z ′ axis) orthogonal to each other, and between any one of the axes and true north. A three-axis electronic compass that calculates the azimuth angle (θ ′) formed in
Magnetic detection means (3, 4, 5) for detecting the geomagnetic vector (H) at an arbitrary measurement position as magnetic data (X, Y, Z) composed of components in three axes, and the dip angle of the geomagnetic vector at the measurement position Means (20) for obtaining (η) and declination (D);
An inclined plane formed by two axes selected from the three axes using the magnetic data (X, Y, Z) and the dip angle (η) (for example, an x′y ′ plane) and a ground plane (for example, a first calculation unit (11) that calculates a solution of a posture angle formed between the xy plane and a predetermined calculation formula;
When there are two candidates (α1, α2) in the posture angle solution obtained by the first calculation unit, an appropriate posture angle is selected from the amount of change in posture angle calculated along the time series. A first process, a second process in which an appropriate attitude angle is selected with reference to the azimuth angle calculated immediately before and the attitude angle used at this time, and the azimuth angle calculated immediately before and A second calculation unit (12) for selecting an appropriate posture angle (α1 or α2) from the third process in which an appropriate posture angle is selected from the ratio of the differential values of the posture angle used at the time,
A third calculation unit (13) for calculating an azimuth angle (θ) with respect to magnetic north from the attitude angle (α1 or α2) selected by the second calculation unit and the magnetic data (X, Y, Z);
And a fourth calculation unit (14) for calculating an azimuth angle (θ ′ = θ−D) with respect to true north by removing a declination angle (D) from the azimuth angle (θ). It is.

  In the present invention, since the azimuth angle can be obtained without using a tilt sensor, a three-axis electronic compass suitable for miniaturization and weight reduction can be provided. Moreover, since the portable terminal equipped with the three-axis electronic compass can detect the azimuth angle with high accuracy regardless of the orientation, the operability of the portable terminal can be improved.

  In the above, it is preferable that one of the two axes forming the inclined plane is parallel to the ground plane.

  In the above means, the x-axis is used to determine the tilt angle (posture angle) between the ground plane and the y-axis, and the y-axis is used to determine the tilt angle (posture angle) between the ground plane and the x-axis. If it is parallel to the ground plane, the correct posture angle can be detected.

  For example, the means for obtaining the dip and declination is preferably obtained from the outside through a relay station in the area of use.

  In the above means, if the mobile terminal is within a relay area where radio waves emitted from the relay station reach, it becomes possible to easily obtain the dip angle and declination data from any location, and the direction from these data The angle can be obtained with high accuracy.

  Further, it is preferable that the first to fourth calculation units are formed by a single calculation unit, and a calculation is performed in each calculation unit in response to a command from the control unit.

Since the number of arithmetic units can be reduced, the size and weight can be reduced.
Specifically, as the first process, a change amount of the immediately previous posture angle and a change amount of the current posture angle are calculated, and a ratio between the change amount of the previous posture angle and the change amount of the current posture angle is calculated. When it is outside the predetermined range, the one with the smaller amount of change among the two solution candidates is selected as an appropriate posture angle.

  The second process calculates an absolute value of a difference between the immediately previous attitude angle and the current attitude angle when the ratio in the first process is within a predetermined range, and the absolute value is predetermined. When the range is exceeded, a value obtained from the same calculation formula as the posture angle used last time is selected as the posture angle.

  Further, the third process determines whether or not the immediately preceding posture angle is within a predetermined range centered on ± 90 ° when the absolute value in the second process is within a predetermined range. If it is out of the range, the sign of the ratio (Y ′ / X ′ (ad)) of the immediately preceding differential value calculated from the magnetic data corresponding to the selected two axes is checked, and the sign is If the sign is positive, select the value obtained from the same formula as the attitude angle used last time as the attitude angle, and if the sign is negative, the value is obtained from a formula different from the one used to calculate the attitude angle used last time. A value to be selected as a posture angle.

  In this case, when the third treatment is within a predetermined range centered on ± 90 ° just before the posture angle, one of the magnetic data corresponding to the selected two axes, The sign of the immediately preceding differential value ratio (Z ′ / X ′ (ad)) calculated from the magnetic data corresponding to the other axes not selected is checked. If the sign is positive, it is used last time. The value obtained from the same calculation formula as the posture angle is selected as the posture angle, and if the sign is negative, the value obtained from the calculation formula different from the calculation formula used to calculate the posture angle used last time is selected as the posture angle. It is possible to have a process to perform.

The present invention also includes an (x′y′z ′) orthogonal coordinate system including three axes (x ′ axis, y ′ axis, z ′ axis) orthogonal to each other, and any one of the axes and true north An azimuth calculation method using a three-axis electronic compass that calculates an azimuth angle (θ ′) formed therebetween,
Detecting a geomagnetic vector at an arbitrary measurement position as magnetic data (X, Y, Z) composed of components in three axial directions;
The dip angle (η) and declination angle (D) of the geomagnetic vector at the measurement position are obtained, and 2 selected from the three axes using the magnetic data (X, Y, Z) and the dip angle (η). Calculating a solution of a posture angle formed between an inclined plane (for example, x′y ′ plane) formed by the axis and a ground plane (for example, xy plane) from a predetermined calculation formula;
When there are two candidates (α1, α2) in the solution of the posture angle obtained in the above step, the first one in which an appropriate posture angle is selected from the amount of change in the posture angle calculated along the time series A second process in which an appropriate posture angle is selected with reference to the processing, the azimuth angle calculated immediately before and the posture angle used at this time, and the azimuth angle calculated immediately before and used at this time. A step of selecting one appropriate posture angle (α1 or α2) from the third process in which an appropriate posture angle is selected from the ratio of the differential values of the posture angle.
Calculating an azimuth angle (θ) with respect to magnetic north from the attitude angle (α1 or α2) selected in the step and the magnetic data (X, Y, Z);
And calculating the azimuth angle (θ ′ = θ−D) with respect to true north by removing the declination angle (D) from the azimuth angle (θ).

  In the present invention, since the attitude angle is obtained from the geomagnetic vector on the coordinate plane including the z axis and the output of the magnetic sensor, the inclination angle generated in the three-axis electronic compass can be corrected using the attitude angle. Thus, it is possible to eliminate the need for a tilt sensor for determining the tilt angle as in the prior art. Therefore, a triaxial electronic compass suitable for miniaturization and weight reduction can be obtained.

  In addition, the azimuth angle can be detected with high accuracy regardless of the posture set without being restricted by the posture angle.

  FIG. 1 is a plan view that two-dimensionally shows the relationship between a mobile terminal equipped with a three-axis electronic compass and an azimuth angle, FIG. 2 is a block diagram showing the configuration of the three-axis electronic compass, and FIG. 3 is the principle of tilt correction. FIG. 4 is a side view of the portable terminal that two-dimensionally shows a state in which the pitch angle α is tilted around the x axis.

  FIG. 1 shows a mobile phone shown as a representative example of the mobile terminal 1. The mobile terminal 1 is equipped with an electronic compass.

  As shown in FIG. 2, the three-axis electronic compass 2 is equipped with three magnetic sensors 3, 4, and 5 as magnetic detection means for detecting the magnetic field strength in the axial direction. The magnetic sensors 3, 4, and 5 are arranged in directions orthogonal to each other, the width direction of the mobile terminal 1 is the x ′ axis, the longitudinal direction of the mobile terminal 1 is the y ′ axis, and the thickness direction of the mobile terminal 1 Is the z ′ axis, the magnetic sensor 3 can detect the intensity of the magnetic field generated in the x ′ axis direction, the magnetic sensor 4 can detect the y ′ axis direction, and the magnetic sensor 5 can detect the strength of the magnetic field generated in the z ′ axis direction. It is said that. Accordingly, the three-axis electronic compass 2 has an x′y′z ′ orthogonal coordinate system formed by the three magnetic sensors 3, 4, 5, and the geomagnetic vector H generated around the earth is It is possible to detect each component.

  As the magnetic sensor constituting the magnetic detection means, for example, an MR (Magneto Resistive) sensor, a Hall element, a fluxgate type magnetic sensor (see Japanese Patent Laid-Open Nos. 9-43322 and 11-118892), and the like are used. be able to.

  As shown in FIG. 2, the three-axis electronic compass 2 is provided with a calculation means 10 and a dip and declination acquisition means 20. The calculation means 10 includes a first calculation unit 11, a second calculation unit 12, a third calculation unit 13, and a fourth calculation unit 14. The functions of the first to fourth arithmetic units 11, 12, 13, 14 and the dip angle and declination obtaining unit 20 will be described later.

  In the following description, a horizontal plane (x′y ′ plane) in which the x ′ axis and the y ′ axis of the x′y′z ′ orthogonal coordinate system that changes according to the attitude of the mobile terminal 1 are parallel to the ground. Xyz orthogonal when the y-axis 'faces true north and the z'-axis perpendicular to both the x'-axis and the y-axis' faces the vertical direction (gravity direction). The coordinate system is used.

  Symbols Hx, Hy, Hz denote the magnitudes of the x-axis component, the y-axis component, and the z-axis component in the xyz orthogonal coordinate system of the geomagnetic vector H detected by the three-axis magnetic sensor mounted on the mobile terminal 1 ( Means the strength of the magnetic field. Reference numeral H ′ indicates a horizontal component when the geomagnetic vector H is projected onto the ground plane (xy plane) and indicates the direction of magnetic north.

  The azimuth angle θ shown in FIGS. 1 and 3 is an angle formed by the reference y ′ axis and magnetic north (the horizontal component H ′ of the geomagnetic vector). The azimuth angle θ ′ is an angle formed by the reference y ′ axis and true north, and is the angle that the triaxial electronic compass of the present invention finally seeks.

  Further, reference symbol α shown in FIG. 4 indicates the y axis (or the ground plane (xy plane)) and the rotated y ′ axis (or x ′) when the mobile terminal 1 is rotated about the x ′ axis (x axis). It means a posture angle (hereinafter referred to as a pitch angle) formed by y ′ plane. The symbol β represents the x axis (or the ground plane (xy plane)) and the rotated x ′ axis (or x′y ′ plane) when the mobile terminal 1 is rotated about the y ′ axis (y axis). Means the posture angle (hereinafter referred to as roll angle).

  Here, the symbol η shown in FIG. 3 is an angle formed by the ground plane (xy plane) and the geomagnetic vector H that cuts through the ground plane, and means a dip angle (downward is a plus). However, the dip angle η is a value that varies depending on the location, and tends to increase as the latitude increases.

  The value of the dip angle η is stored, for example, in a memory means (not shown) corresponding to an arbitrary measurement position, via an artificial satellite that constructs a GPS (Global Positioning System) provided in the mobile terminal 1. It is possible to obtain the current measurement position and obtain the dip angle η corresponding to the current measurement position from the internal memory means. Alternatively, when the mobile terminal 1 is a mobile phone, an area (current measurement position) where the mobile phone is used is determined from the position of the relay station connected during a call or mail, Data regarding the dip angle η can be obtained from the outside of the triaxial electronic compass 2, and the triaxial electronic compass 2 has an dip angle and declination obtaining unit 20 as shown in FIG. 2. .

  First, in the simplest case, that is, when the portable terminal 1 is placed at the center of the xyz orthogonal coordinate system and the pitch angle α and the roll angle β are both α = β = 0 °, detection of the azimuth angle θ with respect to magnetic north is detected. A method will be described. Since the pitch angle α and the roll angle β are both 0 °, the xyz orthogonal coordinate system and the x′y′z ′ orthogonal coordinate system are in agreement.

  At this time, each component of the x'y'z 'orthogonal coordinate system of the geomagnetic vector H detected by the electronic compass is the same as each component of the xyz orthogonal coordinate system. Assuming that Hx, Hy, and Hz, the components Hx, Hy, and Hz can be expressed as the following Equation 1 by using the azimuth angle θ and the dip angle η.

  As shown in FIGS. 1 and 3, the azimuth angle θ is an angle formed by the y ′ axis (in this case, coincident with the y axis) and the horizontal component H ′ of the geomagnetic vector. be able to.

  Next, as shown in FIG. 4, a method for calculating the azimuth angle θ when the pitch angle α occurs in the mobile terminal 1 (note that the roll angle β = 0 °) will be described.

  At this time, assuming that the magnetic data detected by the three magnetic sensors constituting the electronic compass, that is, the magnetic data calculated as the x′y′z ′ orthogonal coordinate system, is X, Y, and Z, the above equation 1 is The magnetic data X, Y, Z and the pitch angle α can be expressed as shown in equation (3).

  Incidentally, when the pitch angle α = 0 ° (the roll angle β is also β = 0 °), the following equation 4 is established by substituting α = 0 ° into the equation 3. This means that the xyz orthogonal coordinate system and the x′y′z ′ orthogonal coordinate system completely coincide.

  When the pitch angle α is generated in the mobile terminal 1, the azimuth angle θ can be expressed as the following equation 5 from the equations 2 and 3.

From the above formula 5, it can be seen that it is necessary to know the pitch angle α in order to obtain the azimuth angle θ.
By the way, in the case of a conventional portable terminal equipped with a tilt sensor, the azimuth angle θ can be obtained by using the pitch angle α detected by the tilt sensor. However, in the case of a configuration having no tilt sensor as in the present invention, the pitch angle α can be obtained by the following means.
From the third row of the above formulas 1 and 3, the following formula 6 is established.

Further, there is a relationship of the following formula 7 between the geomagnetic vector H and the magnetic data X, Y, Z.

  Next, as shown in FIG. 3 and FIG. 4, if the angle formed between the geomagnetic vector component Hyz and the z axis when the geomagnetic vector H is projected onto the yz plane of the xyz orthogonal coordinate system is γa, This can be expressed by Equation 8.

  Substituting Equations 7 and 8 into Equation 6 and solving for α yields the pitch angle α as shown in Equation 9 below.

  Therefore, the azimuth angle θ can be obtained by substituting the pitch angle α obtained in the equation 9 into the equation 5. In the Northern Hemisphere, since the dip angle η is downwardly negative, a minus sign (−) is added at the beginning of the parenthesis in Equation 9, and in the Southern Hemisphere, the dip angle η is upwardly positive, so Is followed by a plus sign (+).

  However, as shown in Equation 9, the pitch angle α has two solutions. Therefore, in order to obtain the correct azimuth angle θ, it is necessary to select an appropriate solution as the pitch angle α from the two solution candidates.

  Here, the two solutions in the northern hemisphere are defined as the following equations 10 and 11 with azimuth angles α1 and α2, respectively.

If the calculated azimuth angles obtained by substituting the pitch angles α1 and α2 into Equation 5 are θ _α1 and θ _α2 , respectively, θ _α1 and θ _α2 are defined as in the following Equations 12 and 13. .

FIG. 5 is a graph showing the relationship between the rotation angle θz about the z axis and magnetic data when the pitch angle α = + 30 ° (constant), that is, the pitch angle α = + 30 ° (constant) and the roll angle β = 0. This shows the relationship with the magnetic data X, Y, Z when the mobile terminal 1 is rotated 360 ° around the z-axis of the absolute coordinate system in the state of ° and the depression angle η = 51 °. FIG. 6 is a graph showing the relationship between the pitch angles α1, α2 calculated based on the magnetic data X, Y, Z in FIG. 5 and the azimuth angles θ _α1 , θ _α2 calculated from the pitch angles α1, α2. FIG. 6 is a graph showing the relationship between the differential values X ′, Y ′, Z ′ [mV / deg] of the angles of the magnetic data X, Y, Z in FIG. 5 and the rotation angle θz of the z axis.

5 to 7, the angle at which the y ′ axis (the longitudinal axis of the mobile terminal 1) coincides with magnetic north (N) is 0 ° (360 °). (N) → East (E) → South (S) → West (W) → Magnetic North (N) In this case, the rotation angle is θz = 0 ° when the rotation angle around the z axis is θz. (360 °) is magnetic north (N), θz = 90 ° is east (E), θz = 180 ° is south (S), and θz = 270 ° is west (W). Means. In FIG. 6, the pitch angle α1 obtained by substituting the magnetic data X, Y, and Z into the equation 10 is shown by a solid line, and the pitch angle α2 obtained by substituting the magnetic data X, Y, Z into the equation 11 is also shown. Is indicated by a one-dot chain line. Further, the azimuth angle θ _α1 calculated by substituting the pitch angle α1 in the equation 12 is indicated by a dotted line, and the azimuth angle θ _α2 calculated by substituting the pitch angle α2 in the equation 13 is indicated by a broken line.

  As shown in FIG. 6, the pitch angle α1 indicates an appropriate pitch angle (α1 = + 30 °) because the rotation angle θz around the z axis is in the range of 0 ° to 90 ° and in the range of 270 ° to 360 °. In other ranges (90 ° to 270 ° range), a sine wave that is not an appropriate pitch angle is shown. The pitch angle α2 indicates an appropriate pitch angle (α2 = + 30 °) when the rotation angle θz around the z axis is in the range of 90 ° to 270 °, and other ranges (0 ° to 90 ° and 270 °). In the range of ~ 360 °, a sine wave that is not an appropriate pitch angle is shown.

  That is, the pitch angle α1 is appropriate among the two solutions of the pitch angle α when the rotation angle θz is in the range of θz = 0 ° to 90 °, and the pitch angle α2 is also appropriate when the rotation angle θz is in the range of θz = 90 ° to 270 °. Similarly, it can be seen that the pitch angle α1 is appropriate in the range of θz = 270 ° to 360 °. In addition, it can be seen that the two solutions α1 and α2 of the pitch angle α are reversed at θz = 90 ° (east) and θz = 270 ° (west). That is, θz = 90 ° (east) is an inflection point P1 where the proper solution reverses from the pitch angle α1 to the pitch angle α2, and similarly θz = 270 ° (west) is the proper solution from the pitch angle α2 to the pitch angle α1. The turning point P2 is reversed.

Therefore, if only an appropriate pitch angle can be selected by dividing each range of the rotation angle θz of the mobile terminal 1, it is possible to always obtain an appropriate azimuth angle θ by calculation. Incidentally, the proper azimuth angle θ is θ _α1 in the range of θz = 0 ° to 90 °, θ _α2 in the range of θz = 90 ° to 270 °, and θ in the range of θz = 270 ° to 360 °. _α1 .

  However, the above is a case where the mobile terminal 1 is continuously rotated 360 ° around the z-axis, and when the mobile terminal 1 is actually used, the user is always expected to rotate the mobile terminal 1 360 °. It is impossible to do. That is, in an actual usage environment, the electronic compass can often only obtain magnetic data X, Y, and Z within a narrow range of θz = 0 ° to 360 °. When X, Y, and Z are substituted into the above formulas 10 and 11 to obtain the pitch angles α1 and α2 of the two solutions, it is possible to determine which pitch angles α1 and α2 are appropriate values. Necessary.

  In the following, a method for determining which of the two solution candidates, pitch angle α1 and α2, is an appropriate value will be described.

FIG. 8 is a flowchart showing a main part of a method for obtaining an appropriate pitch angle α.
In FIG. 8, X, Y, and Z indicate magnetic data that are output values of the x ′ axis, the y ′ axis, and the z ′ axis of the three-axis electronic compass. Α ′ indicates the previously selected pitch angle α, which is stored in the memory means. X ′, Y ′, and Z ′ indicate differential values of the magnetic data X, Y, and Z. In addition, each stage of the flowchart is expressed by using ST codes and numbers as in ST1.

(First step)
In ST1, the outputs of the x'-axis magnetic sensor 3, the y'-axis magnetic sensor 4 and the z'-axis magnetic sensor 5 of the three-axis electronic compass 2 are predetermined as magnetic data X, Y, Z. Acquired at any time based on the sampling period.

(Second step)
In ST2, the first calculation unit 11 acquires the dip angle η and the declination D at the current measurement position using the GPS or relay station, and the magnetic data X, Y, By substituting Z into Equations 10 and 11, the calculated values of the pitch angles α1 and α2 are obtained according to the sampling period, respectively.

(First process of the third step)
In ST3, the second calculation unit 12 compares a plurality of pitch angles α1 obtained in ST2, for example, 20 pitch angles α1, obtains a maximum value α1 _max and a minimum value α1 _min, and The amount of change Δα1 of 20 pitch angles α1 is obtained as the following equation (14).

Similarly, the second calculation unit 12 compares the 20 calculated pitch angles α2 to obtain the maximum value α2 _max and the minimum value α2 _min , and further calculates the change amount Δα2 of the pitch angle α2 at this time. The following equation 15 is obtained.

  As shown in FIG. 6, when the rotation angle θz around the z-axis is in the range of θz = 0 ° to 90 ° and in the range of θz = 270 ° to 360 °, the pitch angle α1 shows a substantially constant value + 30 °. Although the change amount Δα1 is a small value, the change amount Δα2 tends to be a large value because the pitch angle α2 is changed by a sine wave. Therefore, in this range, the value of the ratio Δα1 / Δα2 is usually less than 1/3 times.

  Further, when the rotation angle θz around the z-axis is in the range of θz = 90 ° to 270 °, the pitch angle α2 shows a substantially constant value + 30 °, so the change amount Δα2 is a small value, but the pitch angle α1 is a sine. Since the change is caused by the wave, the change amount Δα1 tends to be a large value. Therefore, in this range, the value of the ratio Δα1 / Δα2 usually exceeds three times.

  On the other hand, when the rotation angle θz around the z-axis is around 0 ° (360 °), the pitch angle α1 corresponds to the vicinity of the valley of the sine wave and shows a substantially constant value, so the rotation angle θz around the z-axis is 180. In the vicinity of °, the pitch angle α2 corresponds to the vicinity of the peak portion of the sine wave and shows a substantially constant value. Therefore, the change amounts Δα1, Δα2 of the pitch angles α1, α2 are both small values. Therefore, the ratio Δα1 / Δα2 is in the range of 1/3 <(Δα1 / Δα2) <3 when the rotation angle θz around the z-axis is around 0 ° (360 °) and 180 °.

  Therefore, in ST4, the ratio Δα1 / Δα2 is obtained from the variation Δα1 and the variation Δα2 obtained in ST3, and the variation ratio Δα1 / Δα2 is in the range of 1/3 <(Δα1 / Δα2) <3. By determining whether or not there is, it is roughly classified whether the rotation angle θz around the z-axis is located near θz = 0 ° (360 °), θz = 180 °, or other than that.

  That is, in the case of “Yes” in ST4, that is, in the case of 1/3 <(Δα1 / Δα2) <3 (when the change rate between the change amount Δα1 and the change amount α2 is small), the rotation angle around the z axis. It is determined that θz is positioned close to θz = 0 ° (360 °) or θz = 180 °. In the case of “No”, that is, when 1/3 <(Δα1 / Δα2) <3 (when the rate of change between the change amount Δα1 and the change amount Δα2 is large), θz = 0 ° (360 °) ) Or θz = 180 °.

  At this time, in the case of the latter “No”, the rotation angle θz around the z-axis is not located in the vicinity of θz = 0 ° (360 °) or θz = 180 °, and therefore, of the pitch angles α1 and α2, One pitch angle α is substantially constant at + 30 °, and it can be determined that the other pitch angle shows a large change.

  Therefore, in ST4-1, Δα1 / Δα2 and 1 are compared in magnitude. At this time, if Δα1 / Δα2 <1 (in the case of “Yes”), it can be determined that the pitch angle α2 changes greatly and the pitch angle α1 shows a substantially constant value (+ 30 °). . Therefore, an appropriate pitch angle (α1 = + 30 °) can be obtained by selecting α1 having a small change amount (ST4-1-1).

  Contrary to the above, when Δα1 / Δα2> 1 (in the case of “No”), it is determined that the pitch angle α1 changes greatly and the pitch angle α2 shows a substantially constant value (+ 30 °). can do. Therefore, in this case, it is possible to obtain an appropriate pitch angle (α2 = + 30 °) by selecting α2 having a small change amount (ST4-1-2).

(Second process of the third step)
On the other hand, in the former case, that is, 1/3 <(Δα1 / Δα2) <3 (in the case of “Yes” in ST4), that is, the rotation angle θz around the z axis is θz = 0 ° (360 °) or θz = When the position is near 180 °, it is not possible to determine which one of the pitch angles α1 and α2 is appropriate, so the second processing from ST5 is further performed.

  As shown in FIG. 6, when the absolute value | α1-α2 | of the difference between the pitch angle α1 and the pitch angle α2 is obtained, the absolute value | α1-α2 | is the rotation angle θz around the z-axis is θz. = 90 ° and θz = 270 °, the minimum value (| α1−α2 | ≈0 °) is shown, and gradually increases as the distance from these angles increases. Θz = 0 ° (360 °) and θz = It can be seen that the maximum value (| α1−α2 | ≈80 °) is shown at 180 °.

  Therefore, in ST5, the absolute value | α1-α2 | is obtained, and the magnitude is compared with, for example, an angle of 70 ° with some margin. At this time, when the absolute value is not in the range of | α1-α2 | <70 ° (in the case of “No”), that is, when the absolute value | α1-α2 | It can be seen that the rotation angle θz is located near θz = 0 ° (360 °) or near θz = 180 °, that is, not located near the inflection point P1 (east) or inflection point P2 (west). .

Therefore, since it can be determined that no reversal of the solution has occurred between the two solutions of the pitch angle α, when the previous pitch angle α is the pitch angle α1 _a calculated from the equation 10, the equation 10 is also obtained this time. It is assumed that the pitch angle α1 _b is calculated using When the previous pitch angle α is the pitch angle α2 _a calculated from the equation 11, the pitch angle α2 _b calculated using the equation 11 is also used (ST5-1).

Note that the pitch angles α1 _a and α1 _b and the pitch angles α2 _a and α2 _b are different from the previous value and the current value in terms of the values calculated by Equation 10 or Equation 11.

(Third process of the third step)
On the other hand, if the determination in ST5 is “Yes”, that is, if the absolute value is in the range of | α1-α2 | <70 °, the following third process is performed.

  When the absolute value is in the range of | α1-α2 | <70 °, the rotation angle θz around the z axis is at a position other than θz = 0 ° or near 180 °, that is, near the inflection point P1 or P2. You can see that it is located. However, it is unclear which of pitch angles α1 and α2 is appropriate. Therefore, the process after ST6 is performed.

  In ST6, the pitch angle α ′ used for the previous calculation is called from a memory means (not shown), and the case is divided depending on whether or not the pitch angle α ′ is at a predetermined angle. That is, in ST6, whether or not the pitch angle α ′ used in the previous calculation is in the range of 60 ° <α ′ <120 ° or −60 ° <α ′ <− 120 °, or is in the other range. In the latter case (“No”), the processes of ST7 and ST8 are performed, and in the former case (“Yes”), the processes of ST9 and ST10 are performed.

  In each process of ST7 and ST9, differential values X ', Y', Z 'of magnetic data X, Y, Z of the electronic compass are used.

  Here, the differential values X ′, Y ′, and Z ′ are, for example, 20 samples of magnetic data X, Y, and Z, and the difference between the current data and the immediately preceding data is treated as an output change amount (slope). In other words, the average value of the change amounts corresponding to 19 values is used as the output differential values X ′, Y ′, and Z ′. For example, the individual data constituting the magnetic data X are X1, X2, X3,..., X20, and X2-X1 = X′1, X3-X2 = X′2, X4-X3 = X′3,. .., X20−X19 = X′19, the differential value X ′ is expressed by the following equation (16).

The same applies to the differential values Y ′ and Z ′.

  In ST7, the differential values X ′ and Y ′ and their ratio Y ′ / X ′ (ad) are calculated from the immediately preceding magnetic data X and Y stored in the memory, and the ratio of the immediately preceding differential value is calculated in ST8. The sign of the product of Y ′ / X ′ (ad) and the ratio of the current differential value (Y ′ / X ′) is examined.

  As shown in FIG. 7, the signs of the differential values X ′ and Y ′ with respect to the rotation angle θz around the z axis when the pitch angle α is α = −30 ° are the first range L1 expressed by 0 ° ≦ θz <90 °. , X ′ is positive, Y ′ is negative, and in the second range L2 indicated by 90 ° ≦ θz <180 °, X ′ and Y ′ are both negative, and third is indicated by 180 ° ≦ θz <270 °. In the range L3, X ′ is negative and Y ′ is positive. In the fourth range L4 indicated by 270 ° ≦ θz <360 °, both X ′ and Y ′ are positive. Therefore, the sign of the differential value ratio Y ′ / X ′ is negative in the first range L1, positive in the second range L2, negative in the third range L3, and the fourth range. It becomes positive in the range L4.

The above tendency is established when the pitch angle α is in the range of −90 ° <α <+ 90 °.
On the other hand, the sign of the differential value ratio Y ′ / X ′ in the range where the pitch angle α is −90 °> α and α> + 90 ° (90 ° <α <270 °) is positive in the first range L1. The second range L2 is negative, the third range L3 is positive, and the fourth range L4 is negative.

  Therefore, the rotation angle θz around the z-axis is divided into four regions indicated by the first to fourth ranges L1 to L4 every 90 °, and when the previous pitch angle α ′ is shifted to the current pitch angle α. By examining the sign of the product, it is possible to detect whether or not there is a movement between regions at the rotation angle θz around the z axis. In addition, the movement between areas | regions here is the case where the rotation angle (theta) z around z-axis moved to the 2nd range L2 this time when it was located in the 1st range L1 last time, for example. Say.

  For example, when the product of the immediately preceding differential value ratio Y ′ / X ′ (ad) and the current differential value ratio Y ′ / X ′ is obtained in ST8, the sign of the product is positive. Then, it is determined that there is no movement between regions at the rotation angle θz around the z axis, and a process of selecting the same type of pitch angle α as the previous pitch angle α ′ is performed (ST8-1). That is, when the previous pitch angle α ′ is the pitch angle α1 calculated by the formula 10, the pitch angle α1 calculated by using the formula 10 can also be set this time (the numerical value is different). Is the pitch angle α2 calculated using Equation 11, the pitch angle α2 calculated using Equation 11 (the numerical values are different) can be used.

  On the other hand, when the sign of the product is negative, it is determined that the movement between the regions has occurred at the rotation angle θz around the z axis, and processing for selecting a pitch angle α of a type different from the immediately preceding pitch angle α ′ is performed. (ST8-2). That is, when the previous pitch angle α ′ is the pitch angle α1 calculated from the equation 10, the pitch angle α2 calculated using the equation 11 is selected as an appropriate value this time. Alternatively, when the previous pitch angle α ′ is the pitch angle α2 calculated from Equation 11, the pitch angle α1 calculated using Equation 10 is selected as an appropriate value this time.

9 to 11 are graphs corresponding to FIGS. 5 to 7 when the pitch angle α = 90 ° (constant), the roll angle β = 0 °, and the depression angle η = 51 °. That is, FIG. 9 shows the relationship with the magnetic data X, Y, Z when the mobile terminal 1 is rotated 360 ° around the z axis of the absolute coordinate system, and FIG. 10 shows the magnetic data X, Y, Z of FIG. FIG. 11 is a graph showing the relationship between the pitch angles α1, α2 calculated based on azimuth and the azimuth angles θ _α1 , θ _α2 calculated from the pitch angles α1, α2, and FIG. 11 is a derivative of the magnetic data X, Y, Z in FIG. It is a graph which shows the relationship between value X ', Y', Z '[mV / deg] and rotation angle (theta) z of az axis.

  As shown in FIG. 9, when the pitch angle α is in the vicinity of α = + 90 ° (the same applies to −90 °), the change in the magnetic data Y becomes small and becomes almost constant. Therefore, as shown in FIG. 11, the immediately preceding differential value Y ′ becomes 0 and the immediately preceding differential value ratio Y ′ / X ′ also becomes 0. Therefore, in the vicinity of the pitch angle α = ± 90 °, Determination based on the sign of the product of the ratio Y ′ / X ′ (ad) of the immediately preceding differential value and the ratio Y ′ / X ′ of the current differential value becomes impossible.

  Therefore, in ST6, the previous pitch angle α ′ is in the range of + 60 ° <α ′ <+ 120 ° and −60 °> α ′> − 120 °, that is, the previous pitch angle (posture angle) α ′ is ± 90 °. When it is within a predetermined range as the center, it is possible to distinguish the current pitch angle α1 and α2 by using the differential value ratio Z ′ / X ′ instead of the differential value ratio Y ′ / X ′. It is said.

  That is, in ST9, the differential values X ′ and Z ′ and the differential value ratio Z ′ / X ′ (ad) are calculated from the immediately preceding magnetic data X and Z stored in the memory means (not shown), and the ratio Z ′. Let / X ′ (ad) be the ratio of the previous differential value. In ST10, the sign of the product of the immediately preceding differential value ratio Z ′ / X ′ (ad) and the current differential value ratio (Z ′ / X ′) is examined by the same method as in ST8. .

  If the sign of the product at this time is positive, it is determined that there is no movement between regions in the rotation angle θz around the z axis between the previous time and the current time, and the same type as the previous pitch angle α ′ Is selected (ST10-1). On the other hand, when the sign of the product is negative, it is determined that the inter-region movement has occurred at the rotation angle θz around the z axis between the previous time and the current time, and the pitch angle α is different from the previous pitch angle α ′. Is selected (ST10-2). As a result, an appropriate pitch angle α1 or α2 can be selected as described above.

(Fourth process)
In the fourth step, the third calculation unit 13 substitutes the appropriate pitch angle α1 or α2 obtained in each process of the third step into the above Equation 12 or Equation 13, so that an accurate orientation is obtained. The angle θ is calculated.

  The azimuth angle θ obtained by the above method is an angle formed between magnetic north (the horizontal component H ′ of the geomagnetic vector) and the y ′ axis of the mobile terminal, but between the magnetic north and true north (the rotation axis of the earth). Has a declination D. The declination D has a different value for each location. The data on the declination in Japan is owned by the Geographical Survey Institute, and the data is updated once every three months.

  Since the mobile phone equipped with the electronic compass must always communicate with the relay station, the current position where the mobile terminal 1 is used can be determined from the position of the relay station. Alternatively, the current position may be determined from the GPS. As in the case of the above-described dip angle η, the dip angle and declination obtaining unit 20 can obtain data on the declination D that differs for each measurement position by using the relay station of the mobile phone system or the GPS. Is possible.

(Fifth step)
In the fifth step, the fourth calculation unit 14 calculates the following Expression 17 using the azimuth angle θ calculated in the fourth step and the declination D obtained by the above method. A true azimuth angle θ ′ is determined.

  In the above-described embodiment, the case where the roll angle β is 0 ° and only the pitch angle α is generated has been described. However, when the pitch angle α is 0 ° and only the roll angle β is generated, xyz orthogonal coordinates are generated. It is possible to obtain the azimuth angle θ by a similar processing method by simply exchanging the x axis and y axis of the system and further exchanging the x ′ axis and y ′ axis of the x′y′z ′ orthogonal coordinate system.

  As described above, since the posture angle (pitch angle α or roll angle β) can be obtained without using an inclination sensor in the present invention, it is possible to provide a three-axis electronic compass excellent in miniaturization and weight reduction. it can.

  Further, in the present invention, an appropriate pitch angle can always be selected from two solutions of the pitch angle regardless of the range of the posture angle (pitch angle α or roll angle β). It is possible to detect the azimuth angle θ with respect to and further the azimuth angle θ ′ with respect to true north with high accuracy.

  In addition, when determining the azimuth angle, it is not necessary to turn the mobile terminal equipped with the three-axis electronic compass 360 °, so that the operability of the mobile terminal can be improved.

  In the three-axis electronic compass shown in the above embodiment, the calculation means 10 has been described as having the first to fourth calculation units 11 to 14, but the first to fourth calculation units 11 to 14 are not included. The calculation unit 10 may be formed as one calculation unit 10 and may be operated by each calculation unit in response to a command from a control unit (not shown).

A plan view two-dimensionally showing the relationship between a mobile terminal equipped with a three-axis electronic compass and an azimuth angle; Block diagram showing the configuration of a 3-axis electronic compass, Orientation analysis diagram to explain the principle of tilt correction three-dimensionally, A side view of the mobile terminal that two-dimensionally shows a state in which the pitch angle α is inclined around the x-axis; A graph showing the relationship between the rotation angle around the z axis and the magnetic data when the pitch angle α is + 30 ° (constant); FIG. 5 is a graph showing the relationship between the rotation angle around the z axis and the pitch angle and azimuth angle in the case of FIG. FIG. 5 is a graph showing the relationship between the rotation angle around the z axis and the differential value of magnetic data in the case of FIG. A flowchart showing the main part of the method for obtaining an appropriate pitch angle α, A graph showing the relationship between the rotation angle around the z-axis and magnetic data when the pitch angle α = 90 ° (constant); FIG. 9 is a graph showing the relationship between the rotation angle around the z axis and the pitch angle and azimuth angle in the case of FIG. FIG. 9 is a graph showing the relationship between the rotation angle about the z axis and the differential value of magnetic data in the case of FIG. 9;

Explanation of symbols

1 Mobile terminal (mobile phone)
2 3-axis electronic compass 3, 4, 5 Magnetic sensor (magnetic detection means)
DESCRIPTION OF SYMBOLS 10 Calculation means 11 1st calculation part 12 2nd calculation part 13 3rd calculation part 14 4th calculation part 20 Declination and declination acquisition means D Declination H Geomagnetic vector H 'Horizontal component Hx of geomagnetic vector Geomagnetic vector X-axis component Hy of H y-axis component Hz of the geomagnetic vector H z-axis components P1, P2 of the geomagnetic vector H Inflection points x ′, y ′, z ′ An orthogonal coordinate system (x′y′z ′) fixed to the mobile terminal Cartesian coordinate system)
x, y, z Cartesian coordinate system (coordinate system fixed to the portable terminal when the x′y ′ plane is the ground plane and the z ′ axis is the vertical direction)
X, Y, Z Magnetic data (magnetic sensor output)
α Actual pitch angle (posture angle) of mobile devices
α1, α2 Calculation pitch angle (posture angle)
β Roll angle (Attitude angle)
η Depression angle θ Azimuth angle to magnetic north θ ′ Azimuth angle to true north

Claims (13)

A three-axis electronic compass having an orthogonal coordinate system composed of three axes orthogonal to each other and calculating an azimuth angle formed between any of the axes and true north,
Magnetic detection means for detecting a geomagnetic vector at an arbitrary measurement position as magnetic data composed of components in three axial directions; means for acquiring the dip and declination of the geomagnetic vector at the measurement position;
A first calculation method for calculating a solution of a posture angle formed between an inclined plane formed by two axes selected from the three axes using the magnetic data and the dip angle and a ground plane from a predetermined calculation formula An arithmetic unit;
A first process in which an appropriate posture angle is selected from the amount of change in posture angle calculated in time series when there are two candidates for the posture angle solution obtained by the first calculation unit; A second process in which an appropriate attitude angle is selected with reference to the azimuth angle calculated immediately before and the attitude angle used at this time, and the azimuth angle calculated immediately before and the attitude used at this time A second calculation unit that selects one appropriate posture angle from the third process in which an appropriate posture angle is selected from the ratio of the differential values of the angles;
A third computing unit that calculates an azimuth angle with respect to magnetic north from the attitude angle selected by the second computing unit and the magnetic data;
A three-axis electronic compass comprising: a fourth arithmetic unit that calculates an azimuth angle with respect to true north by removing a declination from the azimuth angle.   The triaxial electronic compass according to claim 1, wherein any one of the two axes forming the inclined plane is parallel to the ground plane.   The three-axis electronic compass according to claim 1, wherein the means for obtaining the dip angle and the declination angle is obtained via a relay station in the area of use.   The three-axis electronic compass according to claim 1, wherein the first to fourth calculation units are formed by a single calculation unit, and a calculation is performed in each calculation unit in response to a command from the control unit.   The first process calculates a change amount of the immediately previous posture angle and a change amount of the current posture angle, and a ratio between the change amount of the previous posture angle and the change amount of the current posture angle is out of a predetermined range. 2. The three-axis electronic compass according to claim 1, wherein one of the two solution candidates having a smaller amount of change is selected as an appropriate posture angle.   The second process calculates an absolute value of a difference between the immediately previous attitude angle and the current attitude angle when the ratio in the first process is within a predetermined range, and the absolute value is a predetermined value. The three-axis electronic compass according to claim 1, wherein a value obtained from the same calculation formula as the posture angle used last time is selected as the posture angle when the range is exceeded.   The third process determines whether or not the immediately preceding posture angle is within a predetermined range centered on ± 90 ° when the absolute value in the second process is within a predetermined range; If it is out of the range, the sign of the ratio of the immediately preceding differential value calculated from the magnetic data corresponding to the selected two axes is checked. If the sign is positive, the attitude angle used last time and A value obtained from the same calculation formula is selected as the posture angle, and when the sign is negative, a value obtained from a calculation formula different from the calculation formula used to calculate the posture angle used last time is selected as the posture angle. The three-axis electronic compass according to claim 1, wherein   In the third process, when the immediately previous posture angle is within a predetermined range centered on ± 90 °, one of the magnetic data corresponding to the selected two axes is not selected. Check the sign of the ratio of the previous differential value calculated from the magnetic data corresponding to the other axis, and if the sign is positive, use the posture calculated by the same formula as the posture angle used last time. 8. The method according to claim 7, further comprising a process of selecting a value obtained from a calculation formula different from a calculation formula used to calculate a posture angle used last time as a posture angle when the angle is selected and the sign is negative. 3-axis electronic compass. An azimuth calculation method using a three-axis electronic compass that includes an orthogonal coordinate system including three axes orthogonal to each other and calculates an azimuth angle formed between any one of the axes and true north,
Detecting a geomagnetic vector at an arbitrary measurement position as magnetic data comprising components in three axial directions;
Obtaining the dip and declination of the geomagnetic vector at the measurement position;
Calculating a solution of a posture angle formed between an inclined plane formed by two axes selected from the three axes using the magnetic data and the dip angle and a ground plane from a predetermined calculation formula;
When there are two candidates for the posture angle solution obtained in the above step, the first process in which an appropriate posture angle is selected from the amount of change in posture angle calculated along the time series, and immediately before Second processing for selecting an appropriate posture angle with reference to the calculated azimuth angle and the posture angle used at this time, and the azimuth angle calculated immediately before and the differentiation of the posture angle used at this time A step of selecting one appropriate posture angle from the third process in which an appropriate posture angle is selected from the ratio of the values;
A step of calculating an azimuth angle with respect to magnetic north from the attitude angle selected in the step and the magnetic data;
And a step of calculating an azimuth angle with respect to true north by removing a declination from the azimuth angle, and an azimuth calculation method using a three-axis electronic compass.   The first process calculates a change amount of the immediately previous posture angle and a change amount of the current posture angle, and a ratio between the change amount of the previous posture angle and the change amount of the current posture angle is out of a predetermined range. 10. The direction calculation method using a three-axis electronic compass according to claim 9, wherein the one with the smaller amount of change among the two solution candidates is selected as an appropriate posture angle.   The second process calculates an absolute value of a difference between the immediately previous attitude angle and the current attitude angle when the ratio in the first process is within a predetermined range, and the absolute value is a predetermined value. The direction calculation method using a three-axis electronic compass according to claim 9, wherein, when the range is exceeded, a value obtained from the same calculation formula as the posture angle used last time is selected as the posture angle.   The third process determines whether or not the immediately preceding posture angle is within a predetermined range centered on ± 90 ° when the absolute value in the second process is within a predetermined range; If it is out of the range, the sign of the ratio of the immediately preceding differential value calculated from the magnetic data corresponding to the selected two axes is checked. If the sign is positive, the attitude angle used last time and A value obtained from the same calculation formula is selected as the posture angle, and when the sign is negative, a value obtained from a calculation formula different from the calculation formula used to calculate the posture angle used last time is selected as the posture angle. The azimuth calculation method using the three-axis electronic compass according to claim 9.   In the third process, when the immediately previous posture angle is within a predetermined range centered on ± 90 °, one of the magnetic data corresponding to the selected two axes is not selected. Check the sign of the ratio of the previous differential value calculated from the magnetic data corresponding to the other axis, and if the sign is positive, use the posture calculated by the same formula as the posture angle used last time. 13. The method according to claim 12, further comprising a process of selecting a value obtained from a calculation formula different from a calculation formula used to calculate a posture angle used last time as a posture angle when the sign is negative and the sign is negative. A direction calculation method using a three-axis electronic compass.