RU2503923C1 - Method to calibrate magnetic compass of pedestrian - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: in environs of pedestrian activity they select and mark a L-shaped route of calibration, made in the form of two horizontal rectilinear orthogonal sections, in the first section they provide for good accuracy of positioning for a global navigation system. During calibration the pedestrian moves in direct and reverse direction in each section of the route, they calculate magnetic azimuths A_m of pedestrian movement by a magnetic compass, they calculate true azimuths of sections A_u: in the first section by signals of the global navigation system, and in the second one - according to its orthogonality to the first section, they calculate differences of the specified magnetic and true azimuths ΔA=A_MA_U during movement in each section of the route, they approximate these differences with a function

which is used during time of pedestrian work for production of the true azimuth

of its movement.

EFFECT: increased accuracy.

4 dwg

Description

The inventive method of calibrating a magnetic compass (MK) of a pedestrian relates to methods for constructing devices for calibrating MK used on moving objects. The method can be used mainly for on-line calibration of a pedestrian’s autonomous navigation system in order to increase the accuracy of determining the bearing azimuth in the absence of signals from global navigation systems (GNS).

It is known that GNS can be used to determine the azimuth of pedestrian movement, whose signals, however, are not always available due to the peculiarities of the location of the pedestrian (mountains, forests, buildings, etc.). Inertial navigation systems, the main component of which is MK, can serve as an alternative to GNS in this case. The ideal case is when the magnetometer has the same sensitivity along the horizontal axes, figure 1, and the achievement of this goal is usually called calibration. In the inventive method, the concept of calibration is considered broader and combined with the elimination of deviation MK.

The accuracy of navigation measurements using magnetometers is associated with distortions of the magnetic field and / or the conditions of its measurement associated with:

1. The mismatch of the Earth’s magnetic axis with the axis of its rotation.

2. Local magnetic anomalies of the territory of natural and technogenic origin.

3. The deviation of the Earth's magnetic field (MPZ) under the influence of metal objects included in the equipment of a pedestrian.

4. The influence of the dynamics of a particular pedestrian on the position of an unstabilized measuring platform with MK.

The distortions of the MPZ are manifested in the incorrect determination of the azimuth of motion, which is calculated as:

Ψ = arctan (H _y / H _x ),

where N _y and N _x are the horizontal components of the MPZ, the calculation of which takes into account the possible slope of the measuring platform MK.

Distortions of the MPZ are usually divided into HARD and SOFT - distortions.

The first, fig.lb, include distortions due to the additive sum of the form:

H _y = sin * A ± B or H _x = cos * A ± C,

where A is the true azimuth of motion, and B and C are the distortions of the local MPZ, leading to a shift in the center of the circle described by the sensitivity diagram (H _y , H _x ) relative to the origin.

SOFT distortion, figs, are described by the expressions:

H _y = D * sinA, D <1 or

which may apply to the H _x component of the MPZ, and these distortions can manifest themselves in various combinations. The graphically indicated distortions are manifested in the fact that instead of a circle, the graph (H _y , H _x ) becomes an ellipse, which at E = 0 is located strictly horizontally or vertically - depending on which of the components of the MPS there is a distortion D <1. At E ≠ 0, the ellipse tilts. The direction of inclination is determined by the error sign E.

It is established that errors of the Hard type are described in the form of one period, and - of the Soft type - of two periods of harmonic oscillation over the entire azimuth determination interval (0-360 degrees).

The presence of both errors leads to the summation of these harmonic functions with different initial phases and amplitudes. Moreover, as the modeling experience has shown, for accurate approximation of the sum of Hard and Soft errors, sometimes harmonics with periods equal to 1/2 and 1/4 of the period of the fundamental frequency should also be used.

There is a method of calibrating a pedestrian MK for the Honeywell HMR3400 device [1], p. 4, which consists in the fact that the components of the magnetic field are constantly measured by a triaxial magnetometer when the MK rotates slowly in a horizontal plane around a vertical axis. This method allows you to bring the sensitivity diagram of MK to a circular shape.

Similar MK calibration methods are used in domestic navigators, which, as a rule, do not have high accuracy requirements.

The disadvantage of this method is the low accuracy associated with the fact that paragraphs 3, 4 of the list of distortions of the magnetic field above are not taken into account. In addition, with plane rotation of the MC, the errors described by expression (1) at E ≠ 0 are not eliminated, and the deviation of the MC is not eliminated.

A known method of calibrating a pedestrian MK [2], which consists in measuring the signals of all MK axes when it rotates slowly around a vertical axis along a conical surface, so that all MK axes change their position. In this case, an additional (in comparison with the previous case) possibility of taking into account the slopes of the measuring platform and using the obtained data for calibration appears. Otherwise, the disadvantages of this method coincide with the previous option.

There is also a method of calibration of pedestrian MK [3], which consists in measuring the signals of triaxial MK and accelerometers with the calculation of the matrix of compensation coefficients. Accelerometers (inclinometers) allow you to determine the current position of the measuring platform in a static position or at slow inclinations of the measuring platform with MK, recalculate using the indicated matrix the readings of the magnetometers and calculate the azimuth.

The disadvantage of this method is the low accuracy in the case when the carrier of the platform is a person. The complex and personal nature of the dynamics of changes in the position of the human body (platform with a magnetometer) while walking makes solving the problem of virtual stabilization of the platform using accelerometers extremely difficult. And the found azimuth is inaccurate.

The closest to the claimed method of calibrating a magnetic compass is [4], which consists in the fact that the signals of all MK axes are constantly measured, as well as its position using GNS consumer equipment, magnetic A _m and true A _u azimuths of the moving object are calculated. The mismatch between them (which is the sum of the MK deviation and the influence of Hard and Soft errors) is used to correct MK readings and to calculate a set of unknown parameters (tilt angles of the measuring platform and Soft and Hard errors), using data on the coordinates of the moving object. The method is intended for use on vehicles and is a way to constantly refine MK according to the STS, i.e. MK calibration during operation.

The disadvantage of this method is its low accuracy, especially when calibrating the MK used when moving a pedestrian, because:

- determination of the difference between the magnetic and true azimuths during the operation of the moving object allows you to subsequently take into account the specified difference, however, does not allow you to compensate for the Soft component of the error, i.e. for a full calibration of the MK requires a special procedure;

- the sensitivity functions used in the method [4], can only be used in the zone of relatively small values of the determined unknowns;

- the method assumes that the determined parameters in the sampling zone are constant, which is not typical for the pedestrian even at an interval of less than a second, especially with regard to the tilt angles of the platform, and since the period of receipt of data from the STS is one second, additional errors in determining the correction are inevitable in azimuth.

The problem solved by the claimed method is to increase the accuracy of the operational calibration of the pedestrian’s autonomous navigation system. The inventive method allows to some extent take into account most of the above sources of measurement errors of the magnetic field.

, которую используют во время работы пешехода для получения истинного азимута

его движения при отсутствии данных глобальной навигационной системы путем соответствующей коррекции измерений магнитного компаса.To solve the problem in the method of calibrating a pedestrian’s magnetic compass, which consists in constantly measuring the signals of all the axes of the magnetic compass, as well as its position using the consumer’s equipment of the global navigation system, the magnetic compass and consumer equipment of the global navigation system are placed on a specific pedestrian equipped with all necessary equipment for work, in the vicinity of the pedestrian's area of activity, a L-shaped calibration track is selected and marked out, in the form of two horizons In the first section, they provide good positioning accuracy for the global navigation system, during calibration, the indicated pedestrian moves forward and backward along each section of the route, magnetic azimuths A _{m of} pedestrian travel along the magnetic compass are calculated, the true azimuths of sections A are calculated _u : in the first section, according to the signals of the global navigation system, and in the second, according to its orthogonality to the first section, the differences of the indicated magnetic and azimuths ΔA = A _m -A _u , when moving on each of the sections of the route, approximate these differences by the function

which is used during pedestrian work to obtain true azimuth

its movement in the absence of data from the global navigation system by appropriate correction of the measurements of the magnetic compass.

Significant differences of the proposed method from the prototype are the following solutions.

The magnetic compass and GNS consumer equipment are placed on a specific pedestrian who has a certain height, walking style, etc., which allows you to take into account the individual properties of the pedestrian during calibration.

In the prototype, the individual characteristics of the pedestrian are not considered.

Calibration is done by a pedestrian equipped with all the necessary equipment for work. The presence of metal objects and their location with respect to measuring equipment in the equipment of a pedestrian can significantly change the distribution of the surrounding magnetic field.

In the prototype, the effect of pedestrian equipment is not considered.

Calibration is performed in the vicinity of the pedestrian's area of activity, which allows you to take into account the influence of local magnetic anomalies.

In the prototype, the issue of calibration localization is not considered.

The shape of the calibration trajectory in the form of a L-shaped trace, consisting of two horizontal rectilinear orthogonal sections, provides a fairly simple, quick selection of the calibration section in the field. The first section should have good positioning accuracy for GNS. The second section should be marked geometrically relative to the first at an angle of 90 ° using available measuring instruments.

In the prototype, the refinement of the testimony of MK according to the STS is made in the process of the moving object. A similar action can be performed in the inventive device, but this does not replace the special calibration procedure on the marked track.

During calibration, the indicated pedestrian moves forward and backward along each section of the route. This approach allows you to create an orthogonal calibration basis, in which the main measurement vectors are easily determined.

In the prototype, a constant comparison of magnetic A _m and true A _u (according to the GNS) azimuths does not allow to obtain such a basis.

- интегральной магнитной невязки, вызванной различными источниками по отношению к истинным значениям азимута A_u. В качестве аппроксимирующей функции могут использоваться гармонические функции, сплайны и другие функции. Аппроксимация гармоническими функциями, при которой требуется выбрать фазу и амплитуду каждой гармоники, является достаточно трудоемкой задачей. При аппроксимации сплайнами вычислительные проблемы сокращаются, а результирующая функция хорошо согласуется с физикой рассматриваемых явлений.The approximation of the difference ΔA = A _m -A _{u of the} magnetic azimuths A _m and the true azimuths of the sections A _u allows us to obtain the analytical dependence

- integral magnetic residual caused by various sources with respect to the true values of the azimuth A _u . As an approximating function, harmonic functions, splines, and other functions can be used. The approximation by harmonic functions, in which it is required to choose the phase and amplitude of each harmonic, is a rather time-consuming task. When approximating by splines, computational problems are reduced, and the resulting function is in good agreement with the physics of the phenomena under consideration.

In the prototype, the least squares approximation is made of the difference between the model and the measured component of the Earth’s magnetic field vector. The model used is significantly less accurate compared to that proposed in the claimed method.

The inventive method is illustrated by the following graphic materials:

Figure 1 is a sensitivity diagram of a magnetic compass, where:

a) lack of distortion;

b) Hard distortion;

c) Soft distortion;

N _x , N _y - horizontal components of the magnetic field, calculated taking into account the slope of the measuring platform with MK.

Figure 2 - calibration track, where:

OL and OM are the sections of the L-shaped track, N is the direction to the true north.

Figure 3 - graphs:

- error values ΔA = A _m -A _u for each section;

.- approximating function

.

Figure 4 - calibration device MK, where:

1 - magnetometers;

2 - interface device;

3 - equipment consumer GNS;

4 - computer unit.

Magnetometers 1 are designed to measure the components of the magnetic field. Such magnetometers are usually integrated in the autonomous navigation board.

The interface device 2 is intended for receiving signals of magnetometers, their amplification, conversion to digital form and transmission to a computer unit 4.

GNS 3 consumer equipment is designed to obtain the coordinates of a pedestrian (measuring platform). The consumer equipment contains its own computer unit, which is not shown in Fig. 4.

Computer unit 4 is designed to process the measurement results and calculate the corrections necessary for the calculation. The computer unit can be implemented on a microcontroller.

Consider the possibility of implementing the proposed method for calibrating a pedestrian MK.

Before starting calibration, a pedestrian who is to work in the zone of interest is equipped with a measuring platform containing MK and GNS consumer equipment, as well as all the necessary equipment for the pedestrian to work as intended. An ordinary pedestrian has as part of his equipment a limited composition of objects that can affect the MPZ. Military personnel, emergency personnel can carry a significant amount of a variety of weapons and tools that can have a significant impact on the MPZ, so calibration is carried out with a full outfit of a pedestrian.

In the vicinity of the proposed pedestrian actions there may be local anomalies of the MPZ, so calibration should be done there. The operational nature of the work does not allow you to select and mark the calibration route with many azimuths, although this could increase the accuracy of the calibration. Under such conditions, a L-shaped section is selected, FIG. 2, containing two orthogonal sections OL and OM. For the first calibration section OL, figure 2, as a rule, you can find a horizontal rectilinear section, which provides good positioning accuracy using GNS. Such a site may be a road, path, field, etc. In this example, the OL section has an angle of 3 ° with a northward direction. The second section OM, figure 2, should be orthogonal to the first. Its marking can be performed by any topographic measuring means. When moving along marked areas of calibration, a pedestrian moves more rectilinearly than the means of measuring MPZ and GNS show. The experiments showed that for satisfactory accuracy the length of the calibration sections should be 70-90 meters.

Calibration is carried out by moving the pedestrian sequentially in sections OL, LO, OM and MO. The abscissas 3 °, 183 °, 93 °, 183 ° correspond to these directions in Fig. 3. In each of the sections, multiple measurements of the components of the magnetic field by triaxial magnetometers 1 are made, Fig. 4, which are amplified, converted into digital form in the interface device 2 and entered into the computer unit 4. For each of the calibration sections, the average magnetic azimuths A _{m of} pedestrian moving along MK. Magnetic measurements are influenced by all of the above factors, so the average value of magnetic azimuth is their integral estimate. At the OL (LO) site, the coordinates of the pedestrian’s location are repeatedly measured using the STS 3 consumer equipment, which are also input into the computer unit 4. Based on the obtained STS data, in the computer unit 4, the true azimuth A _u of the OL (LO) section is calculated, which can be use the average true azimuth in the area. In the second section OM (MO), the true azimuth A _u is determined taking into account the known angle of rotation relative to the first section - 90 °.

, например, кубическим сплайном, фиг.3.The differences ΔA = A _m -A _{u of the} indicated magnetic and true azimuths are calculated, Fig. 3, when moving along each of the sections of the route. Approximate these differences

, for example, cubic spline, figure 3.

используют во время работы пешехода для получения оценки истинного азимута

движения указанного пешехода при отсутствии данных глобальной навигационной системы путем соответствующей коррекции измерений магнитного компаса

Analytical expression

used during pedestrian operation to obtain an estimate of true azimuth

movement of the indicated pedestrian in the absence of data from the global navigation system by appropriate correction of the measurements of the magnetic compass

Thus, the inventive method for calibrating a pedestrian’s magnetic compass is realizable, provides a simple, operational, integrated account of most disturbing factors and improves the accuracy of determining the azimuth of a pedestrian’s movement by the autonomous navigation system for measuring MPZ.

Information sources

1.http: //www51.honevwell.com/aero/common/documents/mvaerospacecatalog-documents/Space-documents/Diqital Compass Solution HMR3400.pdf

2. Patent US 6877237.

3. Patent US 6543146.

4. Patent US 2012004843.

Claims (1)

A method for calibrating a pedestrian’s magnetic compass, which consists in continuously measuring the signals of all the axes of the magnetic compass, as well as its position using consumer equipment of the global navigation system, characterized in that The magnetic compass and consumer equipment of the global navigation system are placed on a specific pedestrian, equipped with all the necessary equipment for work, in the vicinity of the pedestrian's area of activity, a L-shaped calibration path is selected and marked out in the form of two horizontal, straight, orthogonal sections, in the first section they provide good accuracy positioning for the global navigation system, during calibration, the indicated pedestrian moves forward and backward along each section of the route, read magnetic azimuths A_m pedestrian movements on a magnetic compass, calculate the true azimuths of sections A_u: in the first section, according to the signals of the global navigation system, and in the second, by its orthogonality to the first section, the differences of the indicated magnetic and true azimuths ΔА = A are calculated_m-A_u when moving on each of the sections of the route, approximate these differences function

which is used during pedestrian work to obtain true azimuth

 its movement in the absence of data from the global navigation system by appropriate correction of the measurements of the magnetic compass.

Images