U.S. patent application number 10/138425 was filed with the patent office on 2002-11-28 for gps method and apparatus, navigation system, program storage device and computer data signal embodied in carrier wave.
This patent application is currently assigned to PIONEER CORPORATION. Invention is credited to Takemura, Hajime.
Application Number | 20020177952 10/138425 |
Document ID | / |
Family ID | 18984839 |
Filed Date | 2002-11-28 |
United States Patent
Application |
20020177952 |
Kind Code |
A1 |
Takemura, Hajime |
November 28, 2002 |
GPS METHOD AND APPARATUS, NAVIGATION SYSTEM, PROGRAM STORAGE DEVICE
AND COMPUTER DATA SIGNAL EMBODIED IN CARRIER WAVE
Abstract
A GPS apparatus measures each pseudo-range ri from respective
one of received radio waves from n GPS satellites (200i) captured
by a GPS receiver (18) and approximately calculates a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions having such variables as the clock error t
and the coordinates (x, y, z) of the GPS receiver. If a generation
of a multi-path is detected, it approximately calculates the clock
error t and the coordinates as a solution of n+1 simultaneous
equations by performing a convergence calculation with respect to
total n+1 functions obtained by adding one .alpha. fn+1 to the n
functions, where .alpha.fn+1 is obtained by applying a weighting
.alpha. to one function fn+1 having such variables as the clock
error and the coordinates.
Inventors: |
Takemura, Hajime;
(Kawagoe-shi, JP) |
Correspondence
Address: |
MORGAN LEWIS & BOCKIUS LLP
1111 PENNSYLVANIA AVENUE NW
WASHINGTON
DC
20004
US
|
Assignee: |
PIONEER CORPORATION
|
Family ID: |
18984839 |
Appl. No.: |
10/138425 |
Filed: |
May 6, 2002 |
Current U.S.
Class: |
701/478 ;
342/357.25; 342/357.61 |
Current CPC
Class: |
G01S 19/42 20130101;
G01S 19/22 20130101 |
Class at
Publication: |
701/213 ;
342/357.06 |
International
Class: |
G01C 021/00 |
Foreign Application Data
Date |
Code |
Application Number |
May 8, 2001 |
JP |
2001-137795 |
Claims
What is claimed is:
1. A GPS method comprising: a measurement process of measuring each
pseudo-range ri (i=1, 2, . . . , n) from respective one of received
radio waves from n GPS satellites (n is a natural number equal to
or greater than 3) captured by a GPS receiver mounted on a movable
body; a detection process of detecting a generation of a multi-path
of the respective one of the received radio waves; a first
calculation process of approximately calculating a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions fi (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation process of approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha.fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
process of outputting the coordinates (x, y, z) calculated in said
first or second calculation process as current position data, which
indicate a current position of the movable body; and a selection
process of selecting said second calculation process when the
generation of the multi-path is detected by said detection
process.
2. A GPS method according to claim 1, wherein the n functions fi
(x, y, z, t) are as follows: ri'+(.differential.ri/
.differential.x) .DELTA.x+(.differential.ri/ .differential.y)
.DELTA.y+(.differential.ri/ .differential.z)
.DELTA.z+(.differential.ri/ .differential.t) .DELTA.t, where ri' is
an approximate value of ri, the one function fn+1 (x, y, z, t) is
as follows: rn+1'+(.differential.rn+1/ .differential.x)
.DELTA.x+(.differential.rn+1/ .differential.y)
.DELTA.y+(.differential.rn- +1/ .differential.z)
.DELTA.z+(.differential.rn+1/ .differential.t) .DELTA.t, where
rn+1' is an approximate value of rn+1, and the convergence
calculation is a calculation for converging a pseudo-range error
.DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n) within a
predetermined value in the n simultaneous equations in said first
calculation process and a calculation for converging a pseudo-range
error .DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n, n+1)
within a predetermined value in the n+1 simultaneous equations in
said second calculation process.
3. A GPS method according to claim 1, wherein said second
calculation process uses a function fn+1 (x, y, z, t) which
estimates the clock error t, instead of the one function.
4. A GPS method according to claim 1, wherein said first
calculation process calculates a positioning solution (x, y, z, t)
in converging .DELTA.R=A.multidot..DELTA.X within a predetermined
value when a pseudo-range error matrix of (n, 1) type is
.DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn) and a position
error matrix of (4, 1) type is .DELTA.X=(.DELTA.x, .DELTA.y,
.DELTA.z, .DELTA.t) and a partial differential term matrix of (n,
4) type is A, and said second calculation process calculates a
solution (x, y, z, t) in converging .DELTA.R=A.multidot..DELTA.X
within a predetermined value when a pseudo-range error matrix of
(n+1, 1) type, in which the weighting .alpha. is applied to the n+1
th row, is .DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn,
.alpha..multidot..alpha.rn+1) and a position error matrix of (4, 1)
type is .DELTA.X=(.DELTA.x, .DELTA.y, .DELTA.z, .DELTA.t) and a
partial differential term matrix of (n+1, 4) type in which the
weighting .alpha. is applied to the n+1 th row is A.
5. A GPS method according to claim 1, wherein said detection
process detects the generation of the multi-path depending on
whether or not the pseudo-range ri measured by said measurement
process is unstable.
6. A GPS method according to claim 5, wherein said detection
process detects the generation of the multi-path on the basis of a
variance of the pseudo-range ri measured by said measurement
process.
7. A GPS apparatus comprising: a measurement device for measuring
each pseudo-range ri (i=1, 2, . . . , n) from respective one of
received radio waves from n GPS satellites (n is a natural number
equal to or greater than 3) captured by a GPS receiver mounted on a
movable body; a detection device for detecting a generation of a
multi-path in the respective one of the received radio waves; a
first calculation device for approximately calculating a clock
error t and coordinates (x, y, z) of the GPS receiver as a solution
of n simultaneous equations by performing a convergence calculation
with respect to n functions fi (x, y, z, t) having such variables
as the clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation device for approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha.fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
device for outputting the coordinates (x, y, z) calculated in said
first or second calculation device as current position data, which
indicate a current position of the movable body; and a selection
device for selecting said second calculation device when the
generation of the multi-path is detected by said detection
device.
8. A GPS apparatus according to claim 7, wherein the n functions fi
(x, y, z, t) are as follows: ri'+(.differential.ri/
.differential.x) .DELTA.x+(.differential.ri/ .differential.y)
.DELTA.y+(.differential.ri/ .differential.z)
.DELTA.z+(.differential.ri/ .differential.t) .DELTA.t, where ri' is
an approximate value of ri, the one function fn+1 (x, y, z, t) is
as follows: rn+1'+(.differential.rn+1/ .differential.x)
.DELTA.x+(.differential.rn+1/ .differential.y)
.DELTA.y+(.differential.rn- +1/ .differential.z)
.DELTA.z+(.differential.rn+1/ .differential.t) .DELTA.t, where
rn+1' is an approximate value of rn+1, and the convergence
calculation is a calculation for converging a pseudo-range error
.DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n) within a
predetermined value in the n simultaneous equations in said first
calculation device and a calculation for converging a pseudo-range
error .DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n, n+1)
within a predetermined value in the n+1 simultaneous equations in
said second calculation device.
9. A GPS apparatus according to claim 7, wherein said second
calculation device uses a function fn+1 (x, y, z, t) which
estimates the clock error t, instead of the one function.
10. A GPS apparatus according to claim 7, wherein said first
calculation device calculates a positioning solution (x, y, z, t)
in converging .DELTA.R=A.multidot..DELTA.X within a predetermined
value when a pseudo-range error matrix of (n, 1) type is
.DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn) and a position
error matrix of (4, 1) type is .DELTA.X=(.DELTA.x, .DELTA.y,
.DELTA.z, .DELTA.t) and a partial differential term matrix of (n,
4) type is A, and said second calculation device calculates a
solution in converging .DELTA.R=A.multidot..DELTA.X within a
predetermined value when a pseudo-range error matrix of (n+1, 1)
type, in which the weighting .alpha. is applied to the n+1 th row,
is .DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn,
.alpha..multidot..DELTA.rn+1) and a position error matrix of (4, 1)
type is .DELTA.X=(.DELTA.x, .DELTA.y, .DELTA.z, .DELTA.t) and a
partial differential term matrix of (n+1, 4) type in which the
weighting .alpha. is applied to the n+1 th row is A.
11. A GPS apparatus according to claim 7, wherein said detection
device detects the generation of the multi-path depending on
whether or not the pseudo-range ri measured by said measurement
device is unstable.
12. A GPS apparatus according to claim 11, wherein said detection
device detects the generation of the multi-path on the basis of a
variance of the pseudo-range ri measured by said measurement
device.
13. A navigation system comprising: (i) a GPS apparatus comprising:
a measurement device for measuring each pseudo-range ri (i=1, 2, .
. . , n) from respective one of received radio waves from n GPS
satellites (n is a natural number equal to or greater than 3)
captured by a GPS receiver mounted on a movable body; a detection
device for detecting a generation of a multi-path in the respective
one of the received radio waves; a first calculation device for
approximately calculating a clock error t and coordinates (x, y, z)
of the GPS receiver as a solution of n simultaneous equations by
performing a convergence calculation with respect to n functions fi
(x, y, z, t) having such variables as the clock error t and the
coordinates (x, y, z), each of the n functions fi (x, y, z, t)
representing the pseudo-range ri; a second calculation device for
approximately calculating the clock error t and the coordinates (x,
y, z) as a solution of n+1 simultaneous equations by performing a
convergence calculation with respect to total n+1 functions
obtained by adding one .alpha.fn+1 (x, y, z, t) to the n functions
fi (x, y, z, t), where the one .alpha.fn+1 (x, y, z, t) is obtained
by applying a weighting .alpha. (.alpha. is a real number which is
equal to or greater than 1) to one function fn+1 (x, y, z, t)
having such variables as the clock error t and the coordinates (x,
y, z), the one function fn+1 (x, y, z, t) representing a
pseudo-range rn+1 obtained when the center of the earth is regarded
as one GPS satellite; an output device for outputting the
coordinates (x, y, z) calculated in said first or second
calculation device as current position data, which indicate a
current position of the movable body; and a selection device for
selecting said second calculation device when the generation of the
multi-path is detected by said detection device, (ii) the GPS
receiver; and (iii) a display device for displaying on map data the
current position data outputted from said output device in a
predetermined format.
14. A program storage device readable by a computer, tangibly
embodying a program of instructions executable by the computer to
perform method processes of outputting current position data, which
indicate a current position of a movable body, said method
processes comprising: a measurement process of measuring each
pseudo-range ri (i=1, 2, . . . , n) from respective one of received
radio waves from n GPS satellites (n is a natural number equal to
or greater than 3) captured by a GPS receiver mounted on a movable
body; a detection process of detecting a generation of a multi-path
of the respective one of the received radio waves; a first
calculation process of approximately calculating a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions fi (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation process of approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha.fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
process of outputting the coordinates (x, y, z) calculated in said
first or second calculation process as current position data, which
indicate a current position of the movable body; and a selection
process of selecting said second calculation process when the
generation of the multi-path is detected by said detection
process.
15. A computer data signal embodied in a carrier wave and
representing a series of instructions which cause a computer to
perform method processes of outputting current position data, which
indicate a current position of a movable body, said method
processes comprising: a measurement process of measuring each
pseudo-range ri (i=1, 2, . . . , n) from respective one of received
radio waves from n GPS satellites (n is a natural number equal to
or greater than 3) captured by a GPS receiver mounted on a movable
body; a detection process of detecting a generation of a multi-path
of the respective one of the received radio waves; a first
calculation process of approximately calculating a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions fi (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation process of approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha.fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
process of outputting the coordinates (x, y, z) calculated in said
first or second calculation process as current position data, which
indicate a current position of the movable body; and a selection
process of selecting said second calculation process when the
generation of the multi-path is detected by said detection process.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention generally relates to a GPS (Global
Positioning System) method of and a GPS apparatus for measuring or
positioning a current position of a movable body on the basis of
radio waves received from GPS satellites and to a navigation system
including the GPS apparatus. The present invention also relates to
a program storage device and a computer data signal embodiment in a
carrier wave, which allow a computer to function as the GPS
apparatus.
[0003] 2. Description of the Related Art
[0004] In recent years, a GPS measurement has been broadly used in
a navigation system of a movable body such as a vehicle, an
airplane, a ship, or the like. Coordinates of the GPS satellites,
which are the basis of the GPS measurement, are obtained by solving
the Kepler's equation with the system time included in the
transmitted data of the received radio waves i.e., in the down
links, at a plurality of monitor stations on the earth. Then, an
orbit constant (position) of the satellite as obtained in this
manner is transmitted as one portion of the transmitted data of the
transmitted radio wave i.e., the up link, to the corresponding GPS
satellite. As a result, each GPS satellite transmits its own
position information included in the down link data.
[0005] In the case of 3D (3-dimensional) positioning measurement
based on the radio waves received from the GPS satellites in this
kind, it measures a pseudo-range r, which is a distance from a GPS
receiver to each GPS satellite, on the basis of (i) the position
information of each GPS satellite included in the radio wave
received from each GPS satellite and (ii) a time length required
for the received radio wave coming from each GPS satellite to the
GPS receiver, basically as to four GPS satellites. This
pseudo-range r is represented as a four-variable function of a
clock error t and the coordinates (x, y, z) of the GPS receiver.
Therefore, if the four pseudo-ranges r are measured depending on
the radio waves received from the four GPS satellites, the four
functions are obtained, so that the clock error t and the
coordinates (x, y, z) of the GPS receiver can be calculated by
solving the four nonlinear simultaneous equations. Then, the
above-calculated coordinates are regarded as the current position
of a movable body on which the GPS receiver is mounted.
[0006] In fact, however, it often happens that the radio waves from
five or more GPS satellites can be received at the same time, and
thus, a positioning solution of five (or six or more) nonlinear
simultaneous equations for four variables x, y, z and t are
calculated by the least square method on the basis of the five or
more pseudo-ranges r to increase a positioning accuracy.
[0007] By the way, one disadvantage of the GPS measurement is the
generation of measurement error of the pseudo-range by a
multi-path. More specifically, the radio wave from the GPS
satellite may have other paths to go through to the GPS receiver
after reflecting on the surface of a huge building or the like once
or more times, in addition to the path to go direct to the GPS
receiver, especially around the huge building such as a skyscraper,
a high-rise building, or the like. This phenomenon, under which the
radio waves are received at the same time through a plurality of
radio wave paths from the one identical GPS satellite, is called as
a "multi-path". When the multi-path is generated, an error is
generated in the time length required for the received radio waves
coming to the GPS receiver, and on the basis of this error, an
error is generated in the pseudo-range. Then, when the multi-path
is generated, the pseudo-range based on a regular radio wave path
and the pseudo-ranged based on an irregular radio wave path are
irregularly and alternatively measured, so that the arrival time
and the pseudo-range are changed or flickered at a short cycle.
Thus, the accuracy of the GPS measurement deteriorates
extremely.
[0008] Therefore, a device for detecting the generation of the
multi-path is installed, and if the multi-path is generated, such a
correction that the GPS satellite related to the generation of the
multi-path is eliminated from an object of the GPS measurement is
performed, for example. According to this detection and correction
of the multi-path, when the generation of the multi-path is
detected as for the one or more received radio waves, this or these
are eliminated from the measurement object and it is performed the
GPS measurement based on at least four received radio waves.
Moreover, in the case that only three received radio waves are left
as a result of eliminating the radio waves, in each of which the
generation of the multi-path is detected ,from the measurement
object because of a bad environment of receiving radio waves, such
a technique that the GPS measurement is performed by switching the
3D positioning measurement to the 2D positioning measurement has
been developed. More concretely, in the GPS 2D positioning
measurement, the pseudo-range from each GPS satellite to the GPS
receiver is measured by using (i) the time length required for the
received radio wave to reach from each GPS satellite to the GPS
receiver, and (ii) the position information of each GPS satellite,
which is included in the radio wave received from each GPS
satellite, basically about three GPS satellites under the
assumption that their distances from the earth are rarely changed
for a short time.
[0009] However, according to the present inventors' research, the
above-mentioned technique, which detects the generation of the
multi-path and eliminates the error of the pseudo-range, is not
essentially accurate on its detection. Thus, there is a problem
that the radio wave received through the normal radio wave path may
be often erroneously eliminated from the measurement object of the
GPS measurement by misdetection of the multi-path and the accuracy
of the GPS measurement further deteriorates.
SUMMARY OF THE INVENTION
[0010] It is therefore an object of the present invention to
provide a GPS method and a GPS apparatus, which can detect the
generation of the multi-path and correct an error of the
pseudo-range caused by the effect of the multi-path, as well as
reduce the deterioration of the positioning accuracy even when the
generation of the multi-path is erroneously detected, a navigation
system including the GPS apparatus, a program storage device and a
computer data signal embodiment in a carrier wave, which allow a
computer to function as the GPS apparatus.
[0011] The above object of the present invention can be achieved by
a GPS method provided with: a measurement process of measuring each
pseudo-range ri (i=1, 2, . . . , n) from respective one of received
radio waves from n GPS satellites (n is a natural number equal to
or greater than 3) captured by a GPS receiver mounted on a movable
body; a detection process of detecting a generation of a multi-path
of the respective one of the received radio waves; a first
calculation process of approximately calculating a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions fi (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation process of approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha. fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha. fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
process of outputting the coordinates (x, y, z) calculated in the
first or second calculation process as current position data, which
indicate a current position of the movable body; and a selection
process of selecting the second calculation process when the
generation of the multi-path is detected by the detection
process.
[0012] According to the GPS method of the present invention, the
measurement process measures each pseudo-range ri from the
respective one of received radio waves from n GPS satellites
captured by the GPS receiver. This measurement is basically
performed by measuring a time length required for the arrival and
by multiplying this by the velocity of light. Then, the detection
process detects the generation of the multi-path on the received
radio wave on the basis of whether or not the measurement result,
which is supposed to be obtained experientially, experimentally, or
theoretically when the multi-path is generated, is actually
obtained; for example, such a result that the pseudo-range ri, as
measured above, is unstable. Here, if the generation of the
multi-path is not detected by the detection process, the first
calculation process performs the convergence calculation, for
example, that according to Newton method with respect to the n
functions fi (x, y, z, t), each of which represents the
pseudo-range ri. Consequently, it approximately calculates the
clock error t and the coordinates (x, y, z) as a positioning
solution of the n nonlinear simultaneous equations. As a result, it
calculates the coordinates (x, y, z), which indicate the current
position of the movable body, with a high accuracy under the
condition that the multi-path is not generated.
[0013] On the other hand, if the detection process detects the
generation of the multi-path, the selection process selects the
second calculation process. Then, the second calculation process
performs the convergence calculation, for example, that according
to Newton method with respect to the total n+1 functions obtained
by adding one .alpha.fn+1 (x, y, z, t) to the n functions fi (x, y,
z, t), each of which represents the pseudo-range ri. This one
.alpha. fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), which represents a
pseudo-range rn+1 obtained when the center of the earth is regarded
as one GPS satellite. Consequently, it approximately calculates the
clock error t and the coordinates (x, y, z) as a positioning
solution of the n+1 nonlinear simultaneous equations. As a result,
under the condition of the multi-path generation, as the weighting
.alpha. is applied to the terms associated with the pseudo-range
rn+1 supposed to be stabilized in correspondence to the condition
that the earth is regarded as the GPS satellite, the coordinates
(x, y, z) as the positioning solution are stabilized. In the case
that the generation of the multi-path is misdetected despite of
non-generation of the multi-path in the detection process, as the
weighting .alpha. is applied to the terms associated with the
pseudo-range rn+1 supposed to be stabilized in correspondence to
the condition that the earth is regarded as the GPS satellite, the
coordinates (x, y, z) as the positioning solution are stabilized.
Then, the output process outputs the coordinates, which have the
decreased dispersion or deviation arose from an error caused by the
multi-path generation or by the misdetection of it, as the current
position data, which indicate the current position of the movable
body.
[0014] As described above, both when the multi-path is actually
generated and when the generation of the multi-path is misdetected,
it is possible to calculate the coordinates (x, y, z) as the stable
positioning solution without knowing which case is happening.
Especially, considering such a reality that it is difficult to
increase the accuracy in the detection of the multi-path
generation, the present invention is extremely profitable in
practice because it can obtain substantially the stable GPS
measurement results despite of whether or not the detection of the
multi-path generation is performed properly.
[0015] In one aspect of the GPS method, the n functions fi (x, y,
z, t) are as follows: ri'+(.differential.ri/.differential.x)
.DELTA.x+(.differential.ri /.differential.y)
.DELTA.y+(.differential.ri/ .differential.z)
.DELTA.z+(.differential.ri / .differential.t) .DELTA.t, where ri'
is an approximate value of ri, the one function fn+1 (x, y, z, t)
is as follows: rn+1'+(.differential.rn+1/ .differential.x)
.DELTA.x+(.differential.rn+1/ .differential.y)
.DELTA.y+(.differential.rn- +1/ .differential.z)
.DELTA.z+(.differential.rn+1/ .differential.t) .DELTA.t, where
rn+1' is an approximate value of rn+1, and the convergence
calculation is a calculation for converging a pseudo-range error
.DELTA.ri =.vertline.ri-ri' .vertline.(i=1, 2, . . . , n) within a
predetermined value in the n simultaneous equations in the first
calculation process and a calculation for converging a pseudo-range
error .DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n, n+1)
within a predetermined value in the n+1 simultaneous equations in
the second calculation process.
[0016] According to this aspect, it is possible to calculate the
coordinates (x, y, z) as the positioning solution by performing the
calculation of converging the pseudo-range error
.DELTA.ri=.vertline.ri-r- i' .vertline. within the predetermined
value such as a very small value which is, for example, close to 0
relatively easily and quickly by using a computer or the like in
the first and the second calculation processes.
[0017] The predetermined value used for this convergence
calculation is set in correspondence to the desired positioning
accuracy, and it generally approaches 0 with the positioning
accuracy increased and generally departs from 0 with the
positioning accuracy decreased. This kind of predetermined value is
set in advance depending on the desired positioning accuracy
experientially, experimentally, or theoretically.
[0018] In another aspect of the GPS method, the second calculation
process uses a function fn+1 (x, y, z, t) which estimates the clock
error t instead of the one function.
[0019] According to this aspect, if the detection process detects
the generation of the multi-path, the second calculation process
performs the convergence calculation, for example, that according
to Newton method with respect to the total n+1 functions obtained
by adding one function fn+1 (x, y, z, t) which estimates the clock
error t to the n functions fi (x, y, z, t), each of which
represents the pseudo-range ri, after applying the weighting
.alpha., which is equal to or greater than 1, to the one function
fn+1 (x, y, z, t). Consequently, it approximately calculates the
clock error t and the coordinates (x, y, z) as a positioning
solution of the n+1 nonlinear simultaneous equations. As a result,
under the condition of the multi-path generation, as the weighting
.alpha. is applied to the terms associated with the pseudo-range
rn+1 supposed to be stabilized in correspondence to the estimation
of the clock error t, the coordinates (x, y, z) as the positioning
solution are stabilized. In the case that the generation of the
multi-path is misdetected despite of non-generation of the
multi-path in the detection process, as the weighting .alpha. is
applied to the terms associated with the pseudo-range rn+1 supposed
to be stabilized in correspondence to the estimation of the clock
error t, the coordinates (x, y, z) as the positioning solution are
stabilized. Then, the output process outputs the coordinates, which
have the decreased dispersion or deviation arose from an error
caused by the multi-path generation or by the misdetection of it,
as the current position data, which indicate the current position
of the movable body.
[0020] As described above, both when the multi-path is actually
generated and when the generation of the multi-path is misdetected,
it is possible to calculate the coordinates (x, y, z) as the stable
positioning solution without knowing which case is happening.
[0021] Incidentally, it is also possible to calculate the
coordinates (x, y, z) as the positioning solution by solving n+2
nonlinear simultaneous equations using the one function which
estimates the clock error in addition to the one function obtained
under the condition that the center of the earth is regarded as one
GPS satellite, after applying the weighting .alpha. to at least one
of the functions.
[0022] In another aspect of the GPS method, the first calculation
process calculates a positioning solution (x, y, z, t) in
converging .DELTA.R =A.multidot..DELTA.X within a predetermined
value when a pseudo-range error matrix of (n, 1) type is
.DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn) and a position
error matrix of (4, 1) type is .DELTA.X=(.DELTA.x, .DELTA.y,
.DELTA.z, .DELTA.t) and a partial differential term matrix of (n,
4) type is A, and the second calculation process calculates a
solution (x, y, z, t) in converging .DELTA.R=A.multidot..DELTA.X
within a predetermined value when a pseudo-range error matrix of
(n+1, 1), in which the weighting .alpha. is applied to the n+1 th
row, is .DELTA.R =(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn,
.alpha..multidot..DELTA.rn+1) and a position error matrix of (4, 1)
type is .DELTA.X=(.DELTA.x, .DELTA.y, .DELTA.z, .DELTA.t) and a
partial differential term matrix of (n+1, 4) type in which the
weighting .alpha. is applied to the n+1 th row is A
[0023] According to this aspect, it is possible to calculate the
coordinates (x, y, z) as the positioning solution by performing the
calculation of converging .DELTA.R=A.multidot..DELTA.X within the
predetermined value such as a very small value which is, for
example, close to 0 relatively easily and quickly by a matrix
operation by using a computer or the like in the first and the
second calculation processes.
[0024] In another aspect of the GPS method, the detection process
detects the generation of the multi-path depending on whether or
not the pseudo-range ri measured by the measurement process is
unstable.
[0025] According to this aspect, the generation of the multi-path
is detected by the detection process depending on whether or not
the pseudo-range ri measured by the measurement process is
unstable. In other words, if the pseudo-range ri is not unstable
according to a predetermined standard, the detection process
regards the multi-path as the one without the generation, and if
the pseudo-range ri is unstable according to the predetermined
standard, the detection process regards the multi-path as the one
with the generation.
[0026] In this aspect, the detection process may be constructed to
detect the generation of the multi-path on the basis of variance of
the pseudo-range ri measured by the measurement process.
[0027] By constituting in this manner, it is possible to detect the
generation of the multi-path with relatively high accuracy and ease
on the basis of the fact that the variance or dispersion of
measurement values often varies between the captured GPS satellites
when the multi-path is generated.
[0028] Incidentally, the first and the second calculation process
may be performed by, for example, solving n (or n-1 or less)
nonlinear simultaneous equations about the received radio waves
remained after excluding one or more functions fi (x, y, z, t)
corresponding to the received radio wave or waves, in each of which
the multi-path generation is detected, from the operation object of
the first and the second calculation processes. In general, the
positioning solution can not be obtained if the operation object of
the first and the second calculation processes are less than four
functions including (i) the function which estimates the clock
error and (ii) the function which indicates the pseudo-range and
which is obtained when the center of the earth is regarded as the
GPS satellite, so that at least four functions are adopted as the
operation objects of the first and the second calculation
processes.
[0029] The above object of the present invention can be achieved by
a GPS apparatus comprising: a measurement device for measuring each
pseudo-range ri (i=1, 2, . . . , n) from respective one of received
radio waves from n GPS satellites (n is a natural number equal to
or greater than 3) captured by a GPS receiver mounted on a movable
body; a detection device for detecting a generation of a multi-path
in the respective one of the received radio waves; a first
calculation device for approximately calculating a clock error t
and coordinates (x, y, z) of the GPS receiver as a solution of n
simultaneous equations by performing a convergence calculation with
respect to n functions fi (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), each of the n
functions fi (x, y, z, t) representing the pseudo-range ri; a
second calculation device for approximately calculating the clock
error t and the coordinates (x, y, z) as a solution of n+1
simultaneous equations by performing a convergence calculation with
respect to total n+1 functions obtained by adding one .alpha.fn+1
(x, y, z, t) to the n functions fi (x, y, z, t), where the one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), the one function fn+1
(x, y, z, t) representing a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite; an output
device for outputting the coordinates (x, y, z) calculated in the
first or second calculation device as current position data, which
indicate a current position of the movable body; and a selection
device for selecting the second calculation device when the
generation of the multi-path is detected by the detection
device.
[0030] According to the GPS apparatus of the present invention, the
measurement device measures each pseudo-range ri from the
respective one of received radio waves from n GPS satellites
captured by the GPS receiver. Here, if the generation of the
multi-path is not detected by the detection device, the first
calculation device performs the convergence calculation, for
example, that according to Newton method with respect to the n
functions fi (x, y, z, t), each of which represents the
pseudo-range ri. Consequently, it approximately calculates the
clock error t and the coordinates (x, y, z) as a positioning
solution of the n nonlinear simultaneous equations. As a result, it
calculates the coordinates (x, y, z), which indicate the current
position of the movable body, with a high accuracy under the
condition that the multi-path is not generated.
[0031] On the other hand, if the detection device detects the
generation of the multi-path, the selection device selects the
second calculation process. Then, the second calculation device
performs the convergence calculation, for example, that according
to Newton method with respect to the total n+1 functions obtained
by adding one .alpha.fn+1 (x, y, z, t) to the n functions fi (x, y,
z, t), each of which represents the pseudo-range ri. The one
.alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), which represents a
pseudo-range rn+1 obtained when the center of the earth is regarded
as one GPS satellite. As a result, under the condition of the
multi-path generation, as the weighting .alpha. is applied to the
terms associated with the pseudo-range rn+1 supposed to be
stabilized in correspondence to the condition that the earth is
regarded as the GPS satellite, the coordinates (x, y, z) as the
positioning solution are stabilized. In the case that the
generation of the multi-path is misdetected despite of
non-generation of the multi-path in the detection device, as the
weighting .alpha. is applied to the terms associated with the
pseudo-range rn+1 supposed to be stabilized in correspondence to
the condition that the earth is regarded as the GPS satellite, the
coordinates (x, y, z) as the positioning solution are stabilized.
Then, the output device outputs the coordinates, which have the
decreased dispersion or deviation arose from an error caused by the
multi-path generation or by the misdetection of it, as the current
position data, which indicate the current position of the movable
body.
[0032] As described above, both when the multi-path is actually
generated and when the generation of the multi-path is misdetected,
it is possible to calculate the coordinates (x, y, z) as the stable
positioning solution without knowing which case is happening.
[0033] In one aspect of the GPS apparatus, the n functions fi (x,
y, z, t) are as follows: ri'+(.differential.ri/ .differential.x)
.DELTA.x+(.differential.ri/ .differential.y)
.DELTA.y+(.differential.ri/ .differential.z)
.DELTA.z+(.differential.ri/ .differential.t) .DELTA.t, where ri' is
an approximate value of ri, the one function fn+1 (x, y, z, t) is
as follows: rn+1'+(.differential.rn+1/ .differential.x)
.DELTA.x+(.differential.rn+1/ .differential.y)
.DELTA.y+(.differential.rn- +1/ .differential.z)
.DELTA.z+(.differential.rn+1/ .differential.t) .DELTA.t, where
rn+1' is an approximate value of rn+1, and the convergence
calculation is a calculation for converging a pseudo-range error
.DELTA.ri =.vertline.ri-ri' .vertline.(i=1, 2, . . . , n) within a
predetermined value in the n simultaneous equations in the first
calculation device and a calculation for converging a pseudo-range
error .DELTA.ri=.vertline.ri-ri' .vertline.(i=1, 2, . . . , n, n+1)
within a predetermined value in the n+1 simultaneous equations in
the second calculation device.
[0034] According to this aspect, it is possible to calculate the
coordinates (x, y, z) as the positioning solution by performing the
calculation of converging the pseudo-range error
.DELTA.ri=.vertline.ri-r- i' .vertline.within the predetermined
value such as a very small value which is, for example, close to 0
relatively easily and quickly by using a computer or the like in
the first and the second calculation devices.
[0035] In another aspect of the GPS apparatus, the second
calculation device uses a function fn+1 (x, y, z, t) which
estimates the clock error t instead of the one function.
[0036] According to this aspect, if the detection device detects
the generation of the multi-path, the second calculation device
performs the convergence calculation, for example, that according
to Newton method with respect to the total n+1 functions obtained
by adding one function fn+1 (x, y, z, t) which estimates the clock
error t to the n functions fi (x, y, z, t), each of which
represents the pseudo-range ri, after applying the weighting
.alpha., which is equal to or greater than 1, to the one function
fn+1 (x, y, z, t). As a result, under the condition of the
multi-path generation, as the weighting .alpha. is applied to the
terms associated with the pseudo-range rn+1 supposed to be
stabilized in correspondence to the estimation of the clock error
t, the coordinates (x, y, z) as the positioning solution are
stabilized. In the case that the generation of the multi-path is
misdetected despite of non-generation of the multi-path in the
detection device, as the weighting .alpha. is applied to the terms
associated with the pseudo-range rn+1 supposed to be stabilized in
correspondence to the estimation of the clock error t, the
coordinates (x, y, z) as the positioning solution are stabilized.
Then, the output device outputs the coordinates, which have the
decreased dispersion or deviation arose from an error caused by the
multi-path generation or by the misdetection of it, as the current
position data, which indicate the current position of the movable
body.
[0037] In another aspect of the GPS apparatus, the first
calculation device calculates a positioning solution (x, y, z, t)
in converging .DELTA.R=A.multidot..DELTA.X within a predetermined
value when a pseudo-range error matrix of (n, 1) type is
.DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn), a position
error matrix of (4, 1) type is .DELTA.X=(.DELTA.x, .DELTA.y,
.DELTA.z, .DELTA.t), and a partial differential term matrix of (n,
4) type is A, and the second calculation device calculates a
solution in converging .DELTA.R=A.multidot..DELTA.X within a
predetermined value when a pseudo-range error matrix of (n+1, 1)
type in which the weighting .alpha. is applied to the n+1 th row is
.DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn,
.alpha..multidot..DELTA.rn+1) and a position error matrix of (4, 1)
type is .DELTA.X=(.DELTA.x, .DELTA.y, .DELTA.z, .DELTA.t) and a
partial differential term matrix of (n+1, 4) type in which the
weighting .alpha. is applied to the n+1 th row is A.
[0038] According to this aspect, it is possible to calculate the
coordinates (x, y, z) as the positioning solution by performing the
calculation of converging .DELTA.R=A.multidot..DELTA.X within the
predetermined value such as a very small value which is, for
example, close to 0 relatively easily and quickly by a matrix
operation by using a computer or the like in the first and the
second calculation devices.
[0039] In another aspect of the GPS apparatus, the detection device
detects the generation of the multi-path depending on whether or
not the pseudo-range ri measured by the measurement device is
unstable.
[0040] According to this aspect, the generation of the multi-path
is detected by the detection device depending on whether or not the
pseudo-range ri measured by the measurement device is unstable. In
other words, if the pseudo-range ri is not unstable according to a
predetermined standard, the detection device regards the multi-path
as the one without generation, and if the pseudo-range ri is
unstable according to the predetermined standard, the detection
device regards the multi-path as the one with generation.
[0041] In this aspect, the detection device detects the generation
of the multi-path on the basis of a variance of the pseudo-range ri
measured by the measurement device.
[0042] By constituting in this manner, it is possible to detect the
generation of the multi-path with a relatively high accuracy and
ease on the basis of the fact that the variance or dispersion of
measurement values often varies between the captured GPS satellites
when the multi-path is generated.
[0043] The above object of the present invention can be achieved by
a navigation system provided with: the above-mentioned GPS
apparatus of the present invention (including its various aspects);
the GPS receiver; and a display device for displaying on map data
the current position data outputted from the output device in a
predetermined format.
[0044] According to the navigation system of the present invention,
because it is provided with the above-mentioned GPS apparatus, even
if the multi-path is not generated in practice, even if the
generation of the multi-path is properly detected, and even if the
generation of the multi-path is misdetected, it is possible to
display the current position data on map data on the basis of the
stable positioning results.
[0045] The above object of the present invention can be also
achieved by a program storage device readable by a computer. The
program storage device stores a program of instructions to cause
the computer to function as at least one portion of the
above-described GPS apparatus of the present invention (including
its various aspects).
[0046] According to the program storage device, such as a CD-ROM
(Compact Disc--Read Only Memory), a ROM, a DVD (DVD Read Only
Memory), a floppy disk or the like, of the present invention, the
above described GPS apparatus of the present invention can be
relatively easily realized as a computer reads and executes the
program of instructions or as it executes the program after
downloading the program through communication device. Moreover, the
program of instructions can be sent from a central device with an
application program required for the navigation or other data such
as a map.
[0047] The above object of the present invention can be also
achieved by a computer data signal embodied in a carrier wave and
representing a series of instructions for a computer. The series of
instructions causes the computer to function as at least one
portion of the above-described GPS apparatus of the present
invention (including its various aspects).
[0048] According to the computer data signal embodied in the
carrier wave of the present invention, as the computer downloads
the program in the computer data signal through a computer network
or the like, and executes this program, it is possible to realize
the above described GPS apparatus of the present invention.
[0049] The nature, utility, and further features of this invention
will be more clearly apparent from the following detailed
description with reference to preferred embodiments of the
invention when read in conjunction with the accompanying drawings
briefly described below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0050] FIG. 1 is a block diagram of an on-vehicle navigation system
as a first embodiment of the present invention;
[0051] FIG. 2 is a schematic diagram of GPS measurement in the
embodiment;
[0052] FIG. 3 is one schematic diagram showing a multi-path in the
embodiment;
[0053] FIG. 4 is another schematic diagram showing a multi-path in
the embodiment;
[0054] FIG. 5 is a flow chart showing a GPS method in the first
embodiment; and
[0055] FIG. 6 is a flow chart showing a GPS method in a second
embodiment.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0056] Referring to the accompanying drawings, embodiments of the
present invention will be now explained.
[0057] (I) First Embodiment
[0058] At first, an on-vehicle navigation system in the first
embodiment will be explained with reference to FIG. 1 to FIG.
5.
[0059] As shown in FIG. 1, the on-vehicle navigation system in the
first embodiment is provided with a self-contained positioning
apparatus 10, a GPS receiver 18, a system controller 20, a CD-ROM
drive 31, a DVD-ROM drive 32, a display unit 40, an audio output
unit 50, and an input device 60.
[0060] The self-contained positioning apparatus 10 is constructed
to include an acceleration sensor 11, an angular velocity sensor
12, and a velocity sensor 13. The acceleration sensor 11, which
has, for example, a piezoelectric element, detects an acceleration
of a vehicle and outputs acceleration data. The angular velocity
sensor 12, which has, for example, a vibration gyro, detects an
angular velocity of a vehicle when a direction of the vehicle is
changed and outputs angular velocity data and relative azimuth
data.
[0061] The velocity sensor 13 detects a rotation of a vehicle shaft
mechanically, magnetically or optically and is constructed by a
vehicle speed sensor, which generates a vehicle speed pulse as a
pulse signal at every rotation for a predetermined angle around the
vehicle shaft.
[0062] The GPS receiver 18 is a part to receive a radio wave 19,
which carries the down link data including the data for
positioning, from a plurality of GPS satellites to be used for the
detection of an absolute position of a vehicle by information about
latitude, longitude, or the like.
[0063] The system controller 20 includes an interface 21, a CPU
(Central Processing Unit) 22, a ROM (Read Only Memory) 23 and a RAM
(Random Access Memory) 24 and is constructed to control the whole
navigation apparatus.
[0064] The interface 21 performs an interface operation with the
acceleration sensor 11, the angular velocity sensor 12, the
velocity sensor 13, and the GPS receiver 18. Then from them
respectively, it inputs into the system controller 20 the
acceleration data, the relative azimuth data, the angular velocity
data, the GPS measurement data, the absolute azimuth data etc. in
addition to the vehicle speed pulse. The CPU 22 controls the whole
system controller 20. The ROM 23 has a not-illustrated non-volatile
memory where a control program etc. for controlling the system
controller 20 is stored. The RAM 24 readably stores various types
of data such as route data, which are set in advance by a user
through the input device 60, and supplies a working area to the CPU
22.
[0065] The system controller 20, the CD-ROM drive 31, the DVD-ROM
drive 32, the display unit 40, the audio output unit 50, and the
input device 60 are mutually connected through a bus line 30.
[0066] The CD-ROM drive 31 and the DVD-ROM drive 32 read, under the
control of the system controller 20, a control program
corresponding to each embodiment described below and various types
of data such as road data etc. including the number of traffic
lane, road width, and the like from a CD 33 and a DVD 34,
respectively, and then output them. Incidentally, it is possible to
dispose either one of the CD-ROM drive 31 and the DVD-ROM drive 32,
and it is also possible to dispose a CD and DVD compatible
drive.
[0067] The display unit 40 displays, under the control of the
system controller 20, various types of display data. The display
unit 40 is provided with: a graphic controller 41, which controls
the whole display unit 40 on the basis of control data transmitted
from the CPU 22 through the bus line 30; a buffer memory 42, which
is constructed by a memory such as a VRAM (Video RAM) etc. and
temporarily memorizes immediately displayable image information; a
display control device 43, which controls of display of a display
device 44 such as a liquid crystal device, a CRT (Cathode Ray
Tube), or the like on the basis of image data outputted from the
graphic controller 41; and the display device 44. The display
device 44 is constructed by a liquid crystal display device etc.,
on the order of 5 to 10 inches in diagonal length for example and
is installed around a front panel inside the vehicle.
[0068] The audio output unit 50 is provided with: a D/A (Digital to
Analog) converter 51, which performs a D/A conversion of the sound
digital data transmitted through the bus line 30 from the CD-ROM
drive 31, the DVD-ROM 32, or the RAM 24 etc., under the control of
the system controller 20; an amplifier (AMP) 52, which amplifies a
sound analog signal outputted from the D/A converter 51; and a
speaker 53, which converts the amplified sound analog signal to a
sound and outputs it into the vehicle.
[0069] The input device 60 is provided with a key, a switch, a
button, a remote controller, and so on, to input various types of
commands and data. The input device 60 is installed around the
display device 44 or the front panel of a main body of the
on-vehicle navigation system, which is mounted on the vehicle.
[0070] Next, positional relationships between the GPS satellites
and a vehicle on which the above-constructed on-vehicle navigation
system is mounted and the principle of the GPS measurement will be
explained with reference to FIG. 2 as well as FIG. 1.
[0071] As shown FIG. 2, about twenty GPS satellites 200i are
presently launched into the orbit of the earth 300, and as long as
it is not within tunnels or huge buildings or is not hidden by
them, a GPS receiver 18 can receive, at the same time, the radio
waves 19 which carry the down link data sent from four or more GPS
satellites 200i, in general.
[0072] The radio waves 19, which carry the down link data sent from
the respective GPS satellites 200i, are received by monitor
stations 210 on the earth 300, and are collected to a main control
station 220. Then, the position of each GPS satellite 200i is
obtained by solving Kepler's equation with system time in the down
link data. Then, the obtained orbit constant (position) of each GPS
satellite 200i is transmitted from a transmission station 212 to
each GPS satellite 200i as one portion of a radio wave 29, which
carries the up link data to the corresponding GPS satellite 200i as
well as various information such as a calendar (almanac), a
correction coefficient of a clock, and the like. Each GPS satellite
200i, which receives them, transmits on the radio wave 19 its
position information as well as various information such as a
calendar (almanac), code for measuring distance, and the like,
included in the down link data.
[0073] The GPS receiver 18 mounted on a vehicle 118 is constructed
to receive the radio waves 19, which carry the down link data
including the respective position information from a plurality of
GPS satellites 200i, and to measure the time length required for
the radio wave 19 coming from each GPS satellite 200i to the GPS
receiver 18.
[0074] In FIG. 1 and FIG. 2, especially in the first embodiment,
the following examples are constructed by the system controller 20,
which executes a computer program stored in the CD 33, the DVD 34,
the ROM 23, or the like: one example of the measurement device for
measuring a pseudo-range ri from the respective down link data
carried in the radio waves 19 sent from n GPS satellites 200i (i=1,
2, . . . , n) captured by the GPS receiver 18 mounted on the
vehicle 118; one example of the detection device for detecting the
generation of the multi-path of the radio wave 19; one example of
the first calculation device and one example of the second
calculation device for calculating the coordinates of the vehicle
118 selectively depending on the detection result; and one example
of the output device for outputting the coordinates of the vehicle
118 calculated by the first or the second calculation device as the
current position data showing the current position of the vehicle
118. Here, the first calculation device is intended to calculate a
clock error t and the coordinates (x, y, z) of the current position
of the vehicle 118 approximately as a solution of n simultaneous
equations by performing a convergence calculation according to
Newton method, with respect to n functions fi (x, y, z, t) having
such variables as the clock error t and the coordinates (x, y, z)
of the vehicle 118, each of which represents the pseudo-range ri
when the generation of the multi-path is not detected. On the other
hand, the second calculation device is intended to calculate the
clock error t and the coordinates (x, y, z) of the current position
of the vehicle 118 approximately as a solution of n+1 simultaneous
equations by performing a convergence calculation according to
Newton method with respect to total n+1 functions, when the
generation of the multi-path is detected. These n+1 functions are
obtained by adding one .alpha.fn+1 (x, y, z, t) to n functions fi
(x, y, z, t). This one .alpha.fn+1 (x, y, z, t) is obtained by
applying a weighting .alpha. (.alpha. is a real number which is
equal to or greater than 1) to one function fn+1 (x, y, z, t)
having such variables as the clock error t and the coordinates (x,
y, z), which represents a pseudo-range rn+1 obtained when the
center of the earth is regarded as one GPS satellite 200 n+1.
[0075] Although the positioning solution is calculated by
performing the convergence calculation according to Newton method
in this embodiment, this method is one example to calculate an
approximate solution, so that the other methods can be used for
calculation of the positioning solution of the n simultaneous
equations.
[0076] Here, the multi-path, which is the base of the selection of
whether the first calculation device is used, or whether the second
calculation device is used for the calculation of the current
position coordinates of the vehicle 118 in this embodiment, will be
explained with reference to FIG. 3 to FIG. 4.
[0077] As shown in FIG. 3, the radio wave 19 from the GPS satellite
200i may have a reflected wave 19b, which goes through another path
to the GPS receiver 18 after reflecting once or more times on the
surface of a building 300a, in addition to a direct wave 19a, which
goes direct to the GPS receiver 18, around the building 300a such
as a skyscraper, a high-rise building, or the like. As described
above, when the direct wave 19a and the reflected wave 19b are
received at the same time through a plurality of radio wave paths
from the identical GPS satellite 200i, i.e. when the multi-path is
generated, an error occurs in the time length required for the
direct wave 19a and the reflected wave 19b coming to the GPS
receiver 18 and in the pseudo-range ri based on this error of the
time. For example, if the multi-path is generated, the pseudo-range
based on the direct wave 19a and the pseudo-ranged based on the
reflected wave 19b are irregularly and alternatively measured in
general, so that the arrival time and the pseudo-range ri are
flickered or changed at a short cycle and thus the accuracy of the
GPS measurement deteriorates extremely.
[0078] Theoretically speaking, detecting and eliminating only the
reflected wave 19b are required. However, the fact is that the GPS
receiver 18 receives a composite wave 19c of the direct wave and
the reflected wave as shown in FIG. 4, so that it is difficult to
detect which one is being received in practice. Consequently, when
the multi-path is generated, the detection itself of the multi-path
is relatively difficult, in addition to the difficulty in the
detection and the elimination of only the reflected wave 19b.
[0079] For example, in this embodiment, if the pseudo-range ri is
not unstable on the basis of the dispersion or deviation of the
pseudo-range ri to be measured, it is considered that the
multi-path is not generated. If the pseudo-range ri is unstable, it
is considered that the multi-path is generated. By constituting in
this manner, it is possible to detect the generation of the
multi-path with a relatively high accuracy and ease on the basis of
the fact that the variance or dispersion of measurement values
often varies between the captured GPS satellites 200i when the
multi-path is generated. However, as evidenced by FIG. 4, it still
has a possibility of misdetection of the multi-path generation.
[0080] Considering the above-mentioned detection characteristic of
the multi-path, the GPS method is performed to obtain the stable
GPS measurement result even when the multi-path is misdetected,
when it is detected properly, or when it is not detected, as
described below, in this embodiment.
[0081] The GPS method in the first embodiment and the second
embodiment to be described below is intended to be executed mainly
in the CPU 22, as shown in FIG. 1, and is executed as one portion
of a main navigation program which controls the whole on-vehicle
navigation system to perform a navigation operation. Therefore,
while executing the main navigation system program, operations
shown in each flow chart of the embodiments are always being
executed. Moreover, the computer program in this kind may be stored
in the ROM 23, the CD-ROM 33 or the DVD-ROM 34 and may be
downloaded into the RAM 24 or the like through a communication
device such as a modem, a cell phone, or the like. Instead of or in
addition to it, map data etc. required for the navigation may be
downloaded.
[0082] Next, the on-vehicle navigation system in the first
embodiment, as constructed above, will be explained with reference
to a flow chart in FIG. 5.
[0083] In FIG. 5, when the GPS receiver 18 receives the radio wave
19 from the GPS satellite 200i (step S10), the pseudo-range ri is
measured. This measurement is basically performed by measuring the
required time for the arrival and by multiplying this by the
velocity of light. Then, on the basis of whether or not the
pseudo-range ri measured by the GPS satellite 200i, which is being
received, is unstable, i.e., on the basis of the variance of the
pseudo-range ri or the like, it is judged whether or not the
multi-path is being generated (step S11).
[0084] As a result of this judgment, if the pseudo-range ri is not
unstable (step S11: NO), it is considered that the multi-path is
not generated, and positioning process continues without any change
(step S13). Here, as the first calculation process, the convergence
calculation is performed according to Newton method with respect to
n functions fi (x, y, z, t), each of which indicates the
pseudo-range ri (step S13). In other words, the clock error t and
the coordinates (x, y, z) are approximately calculated as a
positioning solution of n simultaneous equations. Consequently,
under the condition that the multi-path is not generated, the
coordinates (x, y, z) showing the current position of a movable
body can be calculated with a high accuracy.
[0085] More concretely, if an approximate value of ri is ri', the n
functions fi (x, y, z, t) are as follows:
ri'+(.differential.ri/ .differential.x) .DELTA.x+(.differential.ri/
.differential.y) .DELTA.y +(.differential.ri/ .differential.z)
.DELTA.z+(.differential.ri/ .differential.t) .DELTA.t
[0086] In this case, the convergence calculation is a calculation
for converging a pseudo-range error .DELTA.ri=.vertline.ri-ri'
.vertline.(i=1, 2, . . . , n) within a predetermined value
.epsilon., which is close to 0 (.epsilon. is a value set in advance
experientially, experimentally, theoretically, or simulationally
etc. depending on the required positioning accuracy), in the n
simultaneous equations.
[0087] More specifically, when a pseudo-range error matrix of (n,
1) type is .DELTA.R=(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn) and a
position error matrix of (4, 1) type is .DELTA.X=(.DELTA.x,
.DELTA.y, .DELTA.z, .DELTA.t) and a partial differential term
matrix of (n, 4) type is A, the first calculation process
calculates the positioning solution (x, y, z, t) in converging
.DELTA.R=A.multidot..DELTA.x, as shown in Formula (1) below, within
the predetermined value .epsilon.. 1 ( r1 r2 ri rn ) = ( Ax , 1 Ay
, 1 Az , 1 At , 1 Ax , 2 Ay , 2 Az , 2 At , 2 Ax , i Ay , i Az , i
At , i Ax , n Ay , n Az , n At , n ) ( x y z t ) ( 1 )
[0088] On the other hand, if the pseudo-range ri is unstable as a
result of the judgment in step S11 (step S11: YES), it is
considered that the multi-path is generated, and the positioning
method is changed from the above-mentioned 3D positioning method to
the 2D positioning method using the weighting .alpha. (step S14).
Here, the second calculation process in the 2D positioning method
using the weighting .alpha. performs the convergence calculation
according to Newton method with respect to the total n+1 functions
obtained by adding one .alpha.fn+1 (x, y, z, t) to the n functions
fi (x, y, z, t), each of which represents the pseudo-range ri,
where .alpha.fn+1 (x, y, z, t) is obtained by applying a weighting
.alpha. (.alpha. is a real number which is equal to or greater than
1) to one function fn+1 (x, y, z, t) having such variables as the
clock error t and the coordinates (x, y, z), which represents the
pseudo-range rn+1 obtained when the center of the earth is regarded
as one GPS satellite. Consequently, it approximately calculates the
clock error t and the coordinates (x, y, z) as a positioning
solution of n+1 nonlinear simultaneous equations. Incidentally, the
value of the weighting .alpha. used in step S14 can be
appropriately set equal to or greater than 1, although when this
value is greater than 1, it stabilizes the positioning solution, as
described below, more effectively.
[0089] More concretely, if an approximate value of ri is ri', the n
functions fi (x, y, z, t) are as follows:
ri'+(.differential.ri/ .differential.x) .DELTA.x+(.differential.ri/
.differential.y) .DELTA.y +(.differential.ri/ .differential.z)
.DELTA.z +(.differential.ri/ .differential.t) .DELTA.t
[0090] On the contrary, if an approximate value of rn+1 is rn+1',
the one function fn+1 (x, y, z, t) is as follows:
rn+1'+(.differential.rn+1/ .differential.x)
.DELTA.x+(.differential.rn+1/ .differential.y) .DELTA.y
+(.differential.rn+1/ .differential.z)
.DELTA.z+(.differential.rn+1/ .differential.t) .DELTA.t
[0091] In this case, the convergence calculation is a calculation
for converging the pseudo-range error .DELTA.ri=.vertline.ri-ri'
.vertline.(i=1, 2, . . . , n, n+1) within the predetermined value
.epsilon. in the n+1 simultaneous equations.
[0092] More specifically, when a pseudo-range error matrix of (n+1,
1) type, in which the weighting .alpha. is applied to the n+1 th
row, is .DELTA.R =(.DELTA.r1, .DELTA.r2, . . . , .DELTA.rn,
.alpha..multidot..DELTA.rn+1), a position error matrix of (4, 1)
type is .DELTA.X=(.DELTA.x, .DELTA.y, .DELTA.z, .DELTA.t) and a
partial differential term matrix of (n+1, 4) type is A, in which
the weighting .alpha. is applied to the n+1 th row, the second
calculation process calculates the solution (x, y, z, t) in
converging .DELTA.R=A.multidot..DELTA.X, as shown in Formula (2)
below, within the predetermined value .epsilon.. 2 ( r1 r2 ri rn rn
+ 1 ) = ( Ax , 1 Ay , 1 Az , 1 At , 1 Ax , 2 Ay , 2 Az , 2 At , 2
Ax , i Ay , i Az , i At , i Ax , n Ay , n Az , n At , n Ax , n + 1
Ay , n + 1 Az , n + 1 At , n + 1 ) ( x y z t ) ( 2 )
[0093] Especially in this case, the weighting .alpha. is performed
with respect to the terms, which are placed at the bottom row and
which are added in regarding the center of the earth 300 as one GPS
satellite 200n+1, i.e. with respect to .DELTA.rn+1 in the left-side
and Ax,n+1, Ay,n+1, Az,n+1, At,n+1 in the right-side partial
differential term matrix (or a direction cosine matrix) A.
[0094] Consequently, according to the first embodiment, under the
condition of the multi-path generation, as the weighting .alpha. is
performed with respect to the terms associated with the
pseudo-range rn+1 to be stabilized in correspondence to the
condition that the earth is regarded as the GPS satellite in the
step S14, the coordinates (x, y, z) as the positioning solution are
stabilized. In the case that the generation of the multi-path is
misdetected despite of non-generation of the multi-path, as the
weighting .alpha. is performed with respect to the terms associated
with the pseudo-range rn+1 to be stabilized in correspondence to
the condition that the earth is regarded as the GPS satellite in
the step S14, the coordinates (x, y, z) as the positioning solution
are stabilized. On the contrary, under the condition of the
non-generation of the multi-path, the coordinates (x, y, z) as the
positioning solution are stable in the step S13 without changing to
the 2D positioning method or using special weighting. Then, the
current position coordinates of the vehicle 118, which have the
decreased dispersion or deviation arose from an error caused by the
multi-path generation or by misdetection of it, is displayed as the
current position data on a display map on the display device 44
(refer to FIG. 1).
[0095] (II) Second Embodiment
[0096] Next, the second embodiment of the present invention will be
explained with reference to FIG. 6. FIG. 6 is a flow chart showing
a GPS method in the second embodiment. Incidentally, the hardware
structure in the second embodiment is the same as the one in the
first embodiment shown in FIG. 1. The same steps as those in FIG. 5
carry the same step numerals and the detailed explanations of them
are omitted.
[0097] The GPS method in the second embodiment uses a function fn+1
(x, y, z, t) which estimates the clock error t instead of the one
function in the first embodiment, which is added when the center of
the earth 300 is regarded as one GPS satellite 200n+1. As the
function which estimates the clock error t, it is possible to use
the one used in two satellites positioning method, which is known
to those skilled in the art and the concrete explanations of the
function are omitted.
[0098] In FIG. 6, step S10 to step S13 are performed in the same
manner as the first embodiment.
[0099] Then, if the pseudo-range ri is unstable as a result of the
judgment in step S11 (step S11: YES), it is considered that the
multi-path is generated and it is judged whether or not the clock
error t can be estimated (step S22 ). Here, the term "clock error
t" is a time error of the GPS receiver 18 and the GPS satellite and
is caused by impossibility of complete synchronization of a quarts
clock of the GPS receiver 18 and an atomic clock of the GPS
satellite. It can be estimated by the GPS receiver 18 on the basis
of the down link data depending on a condition of receiving the
radio wave 19 including the down link data. Then, if the clock
error t can not be estimated as a result of this judgment (step
S22: NO), the operational flow branches off to the step S13 and the
above-mentioned positioning process continues without any
change.
[0100] On the other hand, if the clock error t can be estimated as
a result of the judgment in step S22 (step S22: YES), the
positioning method is changed from the above-described 3D
positioning method to the one performed by the time error
estimation using the weighting .alpha.. Here, the second
calculation process in the positioning method performed by the time
error estimation using the weighting .alpha. performs the
convergence calculation according to Newton method with respect to
the total n+1 functions obtained by adding the one .alpha. fn+1 (x,
y, z, t) to the n functions fi (x, y, z, t), each of which
indicates the pseudo-range ri, where .alpha.fn+1 (x, y, z, t) is
obtained by the weighting .alpha. (.alpha. is a real number which
is equal to or greater than 1) to the one function fn+1 (x, y, z,
t) which estimates the clock error (step S24). Consequently, in the
substantially same manner as the first embodiment, it approximately
calculates the clock error t and the coordinates (x, y, z) as a
positioning solution of n+1 nonlinear simultaneous equations, as
shown in the Formula (2). Incidentally, the value of the weighting
.alpha. used in step S24 can be appropriately set equal to or
greater than 1, although when this value is greater than 1, it
stabilizes the positioning solution more effectively.
[0101] Especially in this case, the weighting .alpha. is performed
with respect to the terms which are placed at the bottom row and
which are added in positioning after the estimation of the clock
error t in the above Formula (2), i.e. with respect to .DELTA.rn+1
in the left-side and Ax,n+1, Ay,n+1, Az,n+1, At,n+1 in the
right-side partial differential term matrix (or a direction cosine
matrix) A.
[0102] Consequently, according to the second embodiment, under the
condition of the multi-path generation, as the weighting .alpha. is
applied to the terms associated with the pseudo-range rn+1 to be
stabilized in correspondence to the positioning method performed by
the time error estimation in the step S24, the coordinates (x, y,
z) as the positioning solution are stabilized. In the case that the
generation of the multi-path is misdetected despite of
non-generation of the multi-path, as the weighting .alpha. is
applied to the terms associated with the pseudo-range rn+1 to be
stabilized in correspondence to the positioning method performed by
the time error estimation in the step S24, the coordinates (x, y,
z) as the positioning solution are stabilized. On the contrary,
under the condition of the non-generation of the multi-path, the
coordinates (x, y, z) as the positioning solution are stable in the
step S13 without changing to the positioning method performed by
the time error estimation or using the special the weighting. Then,
the current position coordinates of the vehicle 118, which have the
decreased dispersion or deviation arose from an error caused by the
multi-path generation or by misdetection of it, is displayed as the
current position data on a display map on the display device 44
(refer to FIG. 1).
[0103] Although each embodiment described above is associated with
an on-vehicle navigation system, the GPS method and the GPS
apparatus of the present invention are not limited by this and are
allowed to use as those for providing a stable positioning of the
current position of an arbitrary movable body such as an animal, a
human etc., who moves with the navigation system, as well as an
airplane, a ship, or the like.
[0104] The invention may be embodied in other specific forms
without departing from the spirit or essential characteristics
thereof. The present embodiments are therefore to be considered in
all respects as illustrative and not restrictive, the scope of the
invention being indicated by the appended claims rather than by the
foregoing description and all changes which come within the meaning
and range of equivalency of the claims are therefore intended to be
embraced therein.
[0105] The entire disclosure of Japanese Patent Application No.
2001-137795 filed on May 8, 2001 including the specification,
claims, drawings and summary is incorporated herein by reference in
its entirety.
* * * * *