PROCEDE DE DETERMINATION DE PARAMETRES DE CORRECTION

METHOD AND DEVICE FOR DETERMINING CORRECTION PARAMETERS

Abrégé français

Procédé de détermination de paramètres de correction pour les valeurs mesurées d'un compas magnétique, monté dans un véhicule terrestre à des fins de navigation, indiquant l'azimut a de la direction de déplacement du véhicule, d'un inclinomètre indiquant, par rapport à l'horizon, l'élévation e de la direction de déplacement du véhicule, et d'un oedomètre indiquant la distance parcourue s, procédé selon lequel deux parcours d'essai à navigation visuelle sont effectués dans des directions différentes, entre des points de départ et d'arrivée connus. Les valeurs mesurées (a, e, s) sont remplacées par des valeurs corrigées (a', e', s'), comme suit: a' = a + A + B . sin a + C . cos a, e' = e - A2, s' = rho . s et, à partir de la comparaison vectorielle des valeurs connues de direction et de parcours (a', e', s') avec les valeurs mesurées, on détermine les paramètres de correction ci-après: A, pour la déclinaison et l'erreur de montage du compas dans l'azimut, B, C, pour le magnétisme dur et mou du véhicule, A2, pour l'erreur de montage de l'inclinomètre en élévation, et rho , pour une erreur d'échelle de l'oedomètre.

Abrégé anglais

The invention relates
to a method for determining
correction parameters for
the measured values of a
magnetic compass, which
is built into a land craft for
navigation purposes, and
gives the azimuth a of the
direction of motion of the
vehicle; of a gradiometer
giving the elevation e of
the direction of motion of
the vehicle in relation to the
horizon; and of an odometer,
giving the distance s travelled.
In this method, two visually
navigated test drives are
carried out in different directions between known point of departure and
arrival. The measured values (a, e, s) are replaced by corrected
values (a', e', s') in accordance with the following: a' = a + A + B . sin a +
C . cos a; e' = e - A2; S' = p. The correction parameters are
determined by performing a vectorial comparison of the known direction and
distance values (a', e', s') with the measured values. The
correction parameters are as follows: A for declination and compass mounting
errors in azimuth; B, C for hard and soft magnetic vehicle
magnetism; A2 for mounting errors of the gradiometer in elevation; and p for a
scale error of the odometer.

Revendications

The embodiments of the invention in which an exclusive
property or privilege is claimed are defined as follows:
1. Method for determining correction parameters for measured values of a
magnetic
compass, which is built into a land vehicle for navigation purposes, and gives
the
azimuth 'a' of a direction of motion of the vehicle, of a gradiometer giving
the
elevation 'e' of the direction of motion of the vehicle in relation to the
horizon, and of
an odometer giving a distance 's' traveled; an instantaneous direction vector
of the
vehicle being given by
<IMG>
where L j =direction of motion of the vehicle in a horizontal plane and s
j=distance
interval between two measurement instances 'j' and (j-1), characterized in
that a first
test drive is carried out with visual navigation from a departure point with
known
geographical coordinates to a first destination point also with known
geographical
coordinates,
in that a subsequent test drive with a change in direction is carried out
under
visual navigation to a second destination point with known geographical
coordinates,
in that, during the test drives, the measured values (a j, e j, s j) are
recorded at instances t j
(j=1, . . . N), and corresponding values (a'j, e'j; s'j) are calculated from
the known
geographical coordinates, and in that the calculated direction vectors are
related to the
direction vectors determined by measurement according to:
a'=a+A+B-sin a+C-cos a.
e'=e-A2
s'=p-s
and the correction parameters
A for declination and compass mounting error in azimuth,
B, C for hard and soft magnetic magnetism in the vehicle,
A2 for mounting errors of the gradiometer in elevation, and
p for a scale error of the odometer are determined therefrom.

2. Method according to claim 1, characterized in that the second test drive
takes place
from the first destination point back to the departure point.
3. Method according to claim 1, characterized in that the geographical data of
the
departure point and the destination points are determined using a GPS
measuring
system built into the vehicle.
4. Method according to claim 3, characterized in that, during the test drives,
GPS
coordinate determinations are carried out continuously and the accuracy of the
GPS
coordinate values is determined as a ratio to the distance traveled s, and the
destination point is established on the basis of the ratio obtained.

Description

CA 02255115 2005-03-07
Method and Device for Determining Correction Parameters
Field of the Invention
The invention is related to a method for determining correction parameters for
the
measured values of land craft magnetic compass, gradiometer, and odometer.
Rebated Art
When navigating a land vehicle by means of an electronic compass for
indicating the azimuth, a gradiometer for indicating the elevation or height,
and
an odometer for indicating the distance, errors occur in terms of the
calculated
position. The reasons are as follows:
- the difference between magnetic north and the grid north of the map
used for navigation,
- the geometric differences bet~rveen mounting direction of the compass
or gradiometer and the direction of vehicle travel,
- soft and hard magnetic effects of the vehicle on the compass, and
- scale errors when measuring distance by means of the odometer.
For navigation; the direction of vehi~;le travel, expressed in coordinates of
the
grid north system, is used, which is additionally twisted horizontally about
the
declination in reference to the magnetic north system. The declination can be
found in the tables. However, a t,~rn of the compass coordinate system ,
whose direction is unknown, is superimposed to it in relation to the direction
of vehicle travel.
In case of a land vehicle, the direction of vehicle travel is not given from
the
start by the geometrical or optical construction in terms of the vehicle
chassis,
and it cannot be easily calculated according to the manufacturer's
specifications. It can only be determined empirically from the difference
between actual and calculated direction of travel.
From DE 4.1 25 369 A1, a navigational system mounted on a motor vehicle is
known which contains an earth magnetism sensor as an azimuth sensor. To
compensate for nil reports of this sensor due to the effect of a magnetic
environment, a comparison with additionally obtained GPS [global positioning

CA 02255115 1998-10-OS
-2-
system] navigation data is provided. However, this can serve only to correct
a zero point shift of the coordinate system.
In DE 31 41 439 A1, an azimuth determination device is described with which
the vehicle is aligned exactly to north and east by means of an azimuth sensor
mounted on the vehicle. The deviation detected from the two-component
measuring signals of the azimuth sensor in comparison with the given direction
is compensated for by adjusting the measuring signals with an adjustment
circuit. Thus, the distortion of the output signals based on the residual
magnetism is also corrected by means of zero point shift.
The effects of the wrong alignment of a magnetic compass in terms of the
direction of travel and of hard and soft magnetic fields on the reading
accuracy
are known from marine applications. To compensate the deviation of the
compass, coefficients A, B, C, D, E are defined and determined separately. For
example, A takes a constant nil report into account by mounting the compass
twisted in the longitudinal direction of the ship, B takes into account the
effect
of the ship's longitudinal magnetism, C the effect of the athwartship
magnetism, D the effect of magnetism induced in soft-iron parts, and E an
asymmetry of distribution of iron masses in the ship's hull (A. Heine: Kompaf3
ABC [Compass ABC], published by Verlag Klasing + Co., 1983, pages 43-45).
Since the D and E effects are usually smaller, their coefficients are
neglected
as a rule.
To determine the coefficients, the ship is held at known courses to north,
east,
south and west, and the respective deviation of the compass reading from the
known courses is observed. The individual coefficients are calculated from the
mean of the deviation values in comparison with the selected courses.
The deviations from the distances plotted on the map, which in the case of
land navigation result from an inaccurate measuring of the distance, do not
play a role in marine applications. However, for the autonomous navigation of
land vehicles it is necessary to equip the vehicles not only with a magnetic

CA 02255115 2005-03-07
- _,,
-3-
compass for measuring the direction, but also with an odometer for measuring
the distance and with a gradiometer, which allows the conversion of distances
measured in uphill or downhill travel into values that correspond to the map
level.
To calibrate the compass reading, it is customary to drive in a circle and to
record measured values at several defined angular positions in relation to the
centre of the circle. In some cases, the magnetic bearing indication for the
horizontal planes is obtained with incline sensors. In a digital magnetic
compass (DMC) made by the firm of Leica AG in Heerbrugg, Switzerland, not
only three magnetic-field sensors for the three space coordinates, but also
two
incline sensors for elevation and tank are integrated. Distance measuring,
which together with directional measuring results in the position of the
vehicle, .
can be verified and if need be corrected by means of independently obtained
satellite navigation signals (GPS). Direction measuring, distance measuring
and
position measuring are treated as separate independent systems (Information
brochure published by the firm of KVH Industries, Inc., USA, 1995, TACNAV
. System).
Backctround of the Invention
Object of the invention is to create a simple method with which the correction
parameters for the readings of integrated measuring devices for azimuth,
elevation and distance can be dE;termined in a land vehicle; achieving a
considerably increased accuracy in navigation and m-aki~ng a GPS review
unnecessary.
For the evaluation, it is a special adva stage when the run in both directions
is carried
out between two points whose geographical coordinates are known.

CA 02255115 2005-03-07
-3A
Summar7i of them Invention
The present invention seeks to overcome the disadvantages of the
prior art associated with method and device for determining correction
parameters.
According to one aspect of the invention, a method for determining
correction parameters for measured values of a magnetic compass, which is
built into a land vehicle for navigation purposes, and gives the azimuth 'a'
of a
direction of motion of the vehicle, of a gradiometer giving the elevation 'e'
of
the direction of motion of the vehicle ir~ relation to. the horizon, and of an
odometer giving a distance 's' traveled, an instantaneous direction vector of
the vehicle being given by formula 2 where L~ =direction of motion of the
vehicle in a horizontal plane and s~=distance interval between two
measurement instances 'j' and Q-1), characterized in that a first test drive
is
carried out with visual navigation from a departure point with known
geographical coordinates to a first cestination point also with known
geographical coordinates, in that a subsequent test drive with a change in
direction is carried out under visual navigation to a second destination point
with known geographical coordinates, in that, during the .test drives, the
measured values (a~, e~, s~~ are recorded at instances t~ (j=1, . . . N), and
corresponding values (a'~, e'~, s'~~ are calculated from the known
geographical
coordinates, and in that the calculated direction vectors are related to the
direction vectors determined by measurement according to:
a'=a+A+B~sin a+C~cos a
e,=e-A2
s'=p.s
and the correction parameters.
The "Summary of the Invention" ~~oes not necessarily disclose all the
inventive features. The inventions may reside in a sub-combination of the
disclosed features.

CA 02255115 2005-03-07
-4-
Brief Description of the Drawings
In the drawing, the direction of vehicle travel is shown as a vector of the
horizontal
plane, or map plane, in which
Fig. 1 shows azimuth and elevations, and
Fig. 2 shows the effect.of an az muth error on the elevation readings.
Detailed Description of the Preferred Embodiments
Below, the invention is described by means of an example, whereby azimuth
measurement and elevation measurement are performed with a mutually
coupled system (17MC).
Starting with the values for azimuth and elevation, the direction of vehicle
travel LF is shown as a unit vector in the. horizontal plane and related to
magnetic north. Fig. 1 shows the interrelations; where eF is the elevation
angle
between the actual direction of vehicle travel LF and the horizontal plane,
and
aF is the azimuth angle between magnetic north and the projection of LF upon
the horizontal plane. Multiplying this with the travelled distance SF, the
result
in general representation is the position in the horizontal plane:
cos eF ~ cos a~
SF ~ LF = SF COS eF ., Sln aF
SIt1 eF
In practice, the resulting position is determined starting from a starting
point
by adding many interim values L~ , s~; j =1...N. In marine applications, this
method is known dead-reckoning navigation. It therefore follows, by neglecting
the indices N and F for the achieved position along the direction of vehicle
travel in the horizontal plane
cos e~ ~ cos a~
jL ds ~ ~L~ -sj = ~ cns ej -sin a~ -s~
o i=~ ~ i=~
sin e~

CA 02255115 1998-10-OS
-5-
Other forms of an approximated integration are possible and lie within reach
of a person skilled in the art.
Assuming that correct values are available for elevation e, azimuth a and
travel
s, and that the starting point is known, a precise position can be determined
by evaluating the given sum, or an accurate navigation can be achieved by
inserting values taken from the map for a and s. In fact, however, the values
given by the measuring devices are distorted, as mentioned above and a s
shown in Fig. 2.
Elevation a and azimuth a are measured in the coordinate system of the DMC.
Its x axis must correspond to the direction of vehicle travel. Fig, 2 shows
the
azimuth distortion ~aF of the projection of the direction of vehicle travel LF
in
relation to the projection of the x axis and the difference in elevation a of
the
x axis in relation to elevation eF in the direction of vehicle travel.
According to the invention, the measured values a, e, s are replaced by
corrected values a', e', s' according to the following formula:
a' - a + A + B ~ sin a + C ~ cos a
e' = e-A2
s' =p~s
The parameters take the following into consideration:
A the declination with regard to magnetic north and a compass mounting
error in azimuth.
B, C a hard and soft magnetic vehicle magnetism.
A2 a mounting error of the gradiometer in elevation, and
p a scale error of the odometer.

CA 02255115 1998-10-OS
-6-
According to the invention, the unknown correction parameters are determined
in two test runs with visual navigation, where the corrected values a', e' and
s' between the start and finish positions are known and the actual values a,
e,
s are measured. Each test run results in a sufficient number of independent
conditional equations according to the measured values for the three space
coordinates x, y, z, so that the correction parameters are clearly definable
and
can be taken into consideration in the measuring system during subsequent
runs with instrumental navigation.
It is particularly advantageous to simply reverse the direction in the two
test
runs. In that case, only two points and their geographic coordinates must be
known, and the actual travel difference equals zero, so that the system of the
conditional equations is simplified.
When a vehicle is driven only in the plane, gradiometers are unnecessary. This
reduces the number of correction parameters and also simplifies the solution
of the conditional equations. This specialization does not diminish the
inventive
idea. The important finding is that the corrected azimuth can be described as
indicated and that in the simplest case if two test runs between two known
points, all necessary correction parameters can be determined.

Dessin représentatif