JP2003114272A - Gps receiver utilizing ura for outputting 2drms, 2drms calculating method and car navigation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a GPS receiver capable of adaptably estimating an error in a GPS system, even if reliability of a GPS system is improved or deteriorated, and accurately reflecting that in a 2DRMS. SOLUTION: The GPS receiver using a Kalman filter for calculating a positioning solution uses an estimated error σH- Kalman in a horizontal (latitude, longitude) direction calculated from diagonal elements of an error covariance matrix obtained in a calculating process for an estimated value for the Kalman filter for calculating the 2DRMS in accordance with a definition, 2DRMS =2× (σH- Kalman)<2> +(HDOP×σUERE)<2> }<1/2> and outputting it together with the GPS positioning solution. Herein, σUERE calculates the URA as a 1σvalue for an error based on URA Index contained in a GPS message and σUserSeg as a 1σ value for an error in a user segment out of GPS system segments in accordance with the following definition: σUERE=(URA<2> +σUserSeg<2> )<1/2> .

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a GPS receiver, and more particularly to calculation of 2DRMS as an accuracy evaluation value for a GPS positioning solution.

[0002]

2. Description of the Related Art In an application using a GPS positioning solution such as a navigation system, it is very important to estimate an error included in the GPS positioning solution. This is because the accuracy of the estimation affects the accuracy of the final application output. As a method of obtaining the estimation error of the GPS positioning solution, 2DRMS (Distance Root Mean Square) calculation is known. 2DRMS typically covers 95% of horizontal error in GPS positioning solutions.
Used by applications as an index of the accuracy of positioning solutions. The following Equation 3 (Equation 3) represents a conventional 2D RMS calculation method.

[Equation 3] Here, HDOP (Horizontal DOP) is a DOP (Dilution Of Precisi) that represents an expansion coefficient of a pseudo range error to a positioning solution error.
is a horizontal component of (on) and is calculated by the GPS receiver as a value that depends on the instantaneous arrangement of satellites used for positioning. In addition, the pseudo range means that the GPS receiver is G
It is the distance from the GPS satellite obtained by measuring the distance measurement signal from the PS satellite. σ _UERE is the UERE (User Equivalent
1) about pseudo-range error
It is a statistical value that represents a value (standard deviation) and is treated as a constant common to all satellites.

[0003]

The relationship between the three segments that make up the GPS system and the error is as follows.
This will be described with reference to FIG. The three segments are a control segment (Control Segment), a space segment (Space Segment), and a user segment (Use
r Segment). The control segment is GP
S satellite tracking, satellite orbit tracking, error check of atomic clock on satellite, update of various positioning calculation data transmitted from satellite to user, satellite orbit correction, and other general satellites Management and operation of. The space segment is the GPS satellite itself. 2 GPS satellites
9 satellites (JST: as of May 31, 2001), each G
The PS satellite always broadcasts a ranging signal including a navigation message. User segment is GPS for positioning
It is the receiver itself. As described above, the GPS system is composed of the three segments.

Next, the error budget
Will be described. The error budget is an estimate of the pseudo range error in the GPS system for each error factor. For example, the following document, "Understanding GPS Pr
inciples And Application ； ELLIOT D.KAPLAN ”(Reference 1),“ GPS WORLD NEWS AND APLICATIONS OF THE GLO
BAL POSITIONING SYSTEM; AN ADVANSTAR PUBLICATION "
(Reference 2), "JGPSC; Satellite Positioning System Conference Bulletin"
In (Reference 3), a list of error budgets is prepared. The error budgets described in these documents are summarized in Table 1 below. As shown in Table 1,
The error budget roughly classifies the error factors in the GPS system into three segments, and further classifies them into finer error factors. Since the error budget is an estimated value of the error, there are some differences between documents depending on the estimation method.

[0005]

[Table 1]

Here, in Table 1, "SA ON" is
When SA (Selective Availability), which is a deliberate deterioration of accuracy by the operator of the GPS system, is being executed, “SA OFF” indicates a case where SA is not being executed. As shown in FIG. 3, UERE is an estimated value of the error in all of the space segment, the control segment, and the user segment. On the other hand, URA (User Range Accuracy)
Is an estimated value of the error in the control segment and the user segment. URA is a dynamic value that reflects the error in the control segment and space segment, which changes from moment to moment, and can be obtained from the navigation data from each satellite.

Based on the matters described above with reference to FIG. 3 and Table 1, the problem of 2DRMS according to the conventional equation 3 will be described. By the way, many GPS independent positioning systems currently used for consumer use a Kalman filter for calculating a positioning solution. The main feature of the Kalman filter as a GPS navigation filter is that information such as system state estimation and error covariance matrix is transmitted from the previous positioning calculation to the next positioning calculation, and the system state is tentatively estimated by the information. And updating the estimated value using the difference between the measured value and the estimated value, to converge the system to a steady state. As described above, in the GPS receiver using the Kalman filter, the past system state is deeply related to the current positioning accuracy.

There are two problems in the calculation of 2DRMS shown in the above equation 3. One is that the factor of past system status is not included in the calculation of 2DMRS. the other one is,
That is, the estimation of the error of the positioning solution depends on the instantaneous satellite constellation. In particular, in mobile reception of GPS, the arrangement of receivable satellites changes from moment to moment. For this reason, the 2DRMS calculated by the method of Equation 3 has a weak correlation on the time axis. This property is different from the accuracy of the GPS positioning solution when a navigation filter such as a Kalman filter is applied. Therefore, the error estimation value (2DRMS) defined by the above Equation 3 may not be able to accurately estimate the error of the GPS positioning solution.

Therefore, it can be conceived to calculate 2DRMS in consideration of not only the HDOP but also the estimation error obtained from the Kalman filter. In this specification, the estimated error in the horizontal direction calculated from the diagonal element of the error covariance matrix of the Kalman filter is defined as EHPE (Expected Horizontal Po
sition Error).

A method for calculating 2D RMS in which this EHPE is added will be shown. In the instantaneous filter operation, after updating the system state, the estimation error in the horizontal direction can be calculated from the error covariance matrix of the Kalman filter. It is usually calculated from the diagonal elements of the error covariance matrix. This value is σ
_It is written as _{H_Kalman} . At this time, 2D with EHPE added
The RMS can be obtained by the definition of the following mathematical expression 4 (Equation 4).

[Equation 4] Formula 4 is the RSS (Root Sum Sq) of 2DRMS of Formula 3 and the estimation error calculated from the error covariance matrix of the Kalman filter.
uare) value. By using 2D RMS with this EHPE, past system status can be reflected in the estimation of the current positioning error. According to the calculation method of Equation 4, the divergence / convergence of the Kalman filter and the 2D
The RMS also increases and decreases, and thus the correlation between the 2DRMS and the actual positioning error becomes high.

[0011] However, σ _{UER E} in the equation 4 which is a calculation formula of 2D RMS in which EHPE is added is a constant determined based on the statistical value of the error budget. This σ
_There is a problem with _URE being a constant. An example of the problem of using σ _UARE as a constant is shown below.

As shown in FIG. 3, the SA of the space segment is one of the error factors of UERE. SA is specifically implemented by changing the clock frequency of the GPS satellites, uploading inaccurate satellite almanac parameters, and the like. As shown in Table 1, SA is larger than any error factor. Therefore, when SA is turned off, the actual positioning error is greatly reduced. SA is OFF
If there is no deterioration in accuracy due to SA, and the value of σ _UARE is set to a constant considering the deterioration in accuracy due to SA, σ
According to Equation 4, having _UARE as one of the parameters
The problem of 2DRMS is larger than the actual positioning error.

The present invention has been made in view of the above circumstances. That is, according to the present invention, even if the reliability of the GPS system is improved or deteriorated, the error is adaptively estimated, and the error is accurately reflected in 2DRMS.
It is an object of the present invention to provide a GPS receiver capable of outputting a car navigation system, and further to provide a car navigation system in which the accuracy of the vehicle position estimation in an application program is improved by using such 2DRMS.

[0014]

Therefore, the invention according to claim 1 uses a Kalman filter for the calculation of the GPS positioning solution, and in the calculation process of the HDOP and the Kalman filter estimated value calculated in the positioning calculation process. G that can output each index of the estimation error σ _{H_Kalman} in the horizontal (latitude, longitude) direction calculated from the diagonal elements of the obtained error covariance matrix
In the PS receiver, using each index, 2DRMS as the accuracy evaluation value of the GPS positioning solution is defined by the above formula 1. Calculate with. Here, σ _UERE is URA, which is the 1σ value of the error based on the URA Index included in the GPS message,
Let σ _{UserSeg be} the 1σ value of the error for the user segment of the GPS system segment,
Definition of Equation 2 above, Ask according to. By calculating σ _UARE according to this definition, the latest information of the error estimation for the URA Index, that is, the control segment and the space segment, which is broadcast from each GPS satellite at any time can be reflected in σ _UARE . Thereby, URA
Even if fluctuates, it becomes possible to accurately reflect that in 2DRMS.

URA has the following definition, where URA Index is N: URA = 2 when N <6^{1 + N / 2} URA = 2 when N ≧ 6^N-2 Can be calculated by (Claim 2).

In this case, it is preferable to use the worst value in the URA index for each GPS satellite receiving the GPS signal as the value N as the URA index used for calculating the URA (claim 3).

The invention according to claim 4 is provided with the GPS receiver according to any one of claims 1 to 3 described above, and 2DRMS output from the GPS receiver is used as an accuracy evaluation value for a GPS positioning solution. It is a car navigation system that estimates the vehicle position. 2DRMS in this case
, Which is accurately reflected even if the URA changes, the car navigation system uses the GPS positioning solution,
By using this 2DRMS when comparing other positioning results such as map matching results, the accuracy of final vehicle position estimation can be improved.

According to a fifth aspect of the present invention, a GPS signal is received to calculate a GPS positioning solution, and a Kalman filter is used to calculate the GPS positioning solution. The HDOP and the Kalman filter are calculated in the positioning calculation process. Estimated error σ _{H_Kal} in the horizontal (latitude, longitude) direction calculated from the diagonal elements of the error covariance matrix obtained in the process of calculating the estimated value
_In an apparatus capable of outputting each _man index, 2DRMS as an accuracy evaluation value of a GPS positioning solution is calculated by using each index according to the definitions of Equation 1 (Equation 1) and Equation 2 (Equation 2). This is a method of calculating an accuracy evaluation value for a positioning solution. This calculation method can be used in a GPS receiver or a car navigation system having a GPS receiver.

URA has the URA Index as N, and
Definition, URA = 2 when N <6^{1 + N / 2} URA = 2 when N ≧ 6^N-2 Can be calculated by (Claim 6).

In this case, it is preferable to use the worst value in the URA index for each GPS satellite receiving a GPS signal as the value N as the URA index used for calculating the URA (claim 7).

[0021]

1 is a block diagram showing the configuration of a GPS receiver as an embodiment of the present invention. GP
In the S receiver 50, the GPS signal is received by the antenna 1 and is input to the frequency conversion unit 2 as the RF signal 13. In the frequency conversion unit 2, the RF signal 13 and the signal generated by the synthesizer unit 3 based on the signal from the reference frequency generator 4 are multiplied, whereby the RF signal 13
Is frequency-converted into the IF signal 14. The IF signal 14 is
The signal is converted into a digital signal by the A / D converter 5 and input to each channel 29 in the digital signal processor 30.

Each channel 29 has a function of receiving and processing GPS signals. Here, channel 2
It is assumed that there is a sufficient number 9 (for example, 12 channels) necessary for the positioning calculation.

The digital signal input from the A / D converter 5 to the channel 29 is sent to the carrier correlator 6 to generate the carrier N.
The carrier component is removed by correlation with the signal generated by CO7 (NCO; Numerically Controlled Oscillator), and the result is input to the CA code correlation unit 8. In the CA code correlation unit 8,
The signal input from the carrier correlation unit 6 is correlated with the duplicate code generated by the CA code generation unit 9 for the satellite to be captured. Based on these processes, the digital signal processing unit 30 obtains pseudo range, navigation data, Doppler frequency phase and the like. These data are
It is input to the CPU 11 via the signal line 21 and used for positioning calculation. The memory 12 connected to the CPU 11 stores various data for positioning calculation.

In the CPU 11, pseudo range, navigation data,
A carrier NCO control signal 18 and a code NCO control signal 19 are generated based on the Doppler frequency phase, and these form a loop for continuously capturing the signal from the satellite.

Further, in the CPU 11, various information is generated by performing positioning calculation based on the pseudo range, navigation data, and Doppler frequency phase. The generated information includes latitude, longitude, altitude as positioning positions, time, speed, azimuth, satellite orbit information, and the like. The generated information is output from the CPU 11 as a GPS output message. A car navigation system can be realized by connecting the GPS receiver 50 to the host CPU so that the host CPU receives the GPS output message. Note that the CPU 11 uses the Kalman filter for calculating the GPS positioning solution, and can use σ _{H_Kalman} which is EHPE for calculating 2DRMS.

Next, 2DRMS in the GPS receiver 50
The calculation will be described. As mentioned above, UERE is classified into three segments. CPU of GPS receiver 50
Numeral 11 requests the URA corresponding to the control segment and the space segment from the navigation message at any time.
Regarding the user segment, a constant is determined from the statistical value in the user segment of the error budget.
Σ is the estimated error value for the user segment
Putting a _{UserSeg, σ UERE} the definition of the variance of the additive from the above equation 2 (equation 2), Can be expressed as

The specific calculation method of URA and σ _UserSeg in Expression 2 is as follows. URA has a different value for each GPS satellite and is used to estimate the 1σ value of each pseudorange error. In reality, indexes of 0 to 15 are assigned to the magnitude of the URA value.
This index is for navigation message subframe 1
Is included in the broadcast. Below, this index
It is referred to as the URA Index. From this URA Index, URA (error 1σ
In order to calculate (value), the following equation 5 (equation 5) and equation 6 (equation 6) are generally used. Here, URA In
dex is N and URA is X.

[Equation 5]

[Equation 6]

In equations (5) and (6), the worst value is selected from the URA index for each satellite being received and is used as N. When the worst value is used, the error can be estimated in the worst case when the GPS system may include a factor that deteriorates the accuracy. That is, Equation 5 or Equation 6 and URAInd
X, which is the 1σ value of the error, is obtained from the worst value of ex, and this X is used as URA in Expression 2.

Next, a method of obtaining σ _UserSeg in Expression 2 will be described. This value is determined as a constant using the error budget (Table 1). here,
The content of the error budget (Table 1) is stored in the memory 12 connected to the CPU 11. User segment error factors, ionospheric delay error, troposphere delay error, receiver noise, multipath, and other 1σ
Values are set as σ _I , σ _T , σ _R , σ _M , and σ _O , respectively.
σ _UserSeg is obtained by the definition of the following Expression 7 (Equation 7) from the additivity of variance.

[Equation 7] As described above, the UR obtained by Equation 5, Equation 6, and Equation 7
Using A and sigma _UserSeg, seeking sigma _UERE formula 2, in the same manner as in Equation 4, definition of 2DRMS (Adaptive URA) to the equation 1 (number 1) as an application of the URA, Calculate with.

On the other hand, a navigation system using the GPS receiver 50 is, for example, a DGP based on FM multiplex broadcasting.
When the GPS receiver 50 is configured to be able to receive S (Differential GPS) correction data broadcast and use the DGPS correction data, the GPS receiver 50 is defined by the following formula 8 (Equation 8). 2DRMS may be calculated. Note that σ _UDRE in Equation 8 is DG
It is the 1σ value of the error using the user differential range error (UDRE) included in the PS correction data.

[Equation 8]

FIG. 2 shows the GPS receiver 50 described above.
5 is a flowchart showing the operation of 2DRMS calculation controlled by the CPU 11 of FIG. Here, as an example, 2D
The calculation of the RMS is executed as a subroutine of the positioning calculation operation in the CPU 11. First, in step S1, DGPS correction data is received and D
It is determined whether GPS positioning is in progress. If DGPS positioning is in progress and DGPS correction data is available (S
1: YES), σ _UDRE is used as σ _UERE (S8), that is, 2DRMS is calculated according to the definition of the above-mentioned expression 8 (S7).

On the other hand, if DGPS positioning is not in progress (S
1: NO), the process proceeds to step S2, and σ _UserSeg is calculated according to the above formula 7. next,
In step S3, URA In in the acquired navigation message
If the dex value is confirmed and the URA Index is smaller than 6 (S
3: YES), and the URA is calculated according to Equation 5 (S4). When the URA Index is 6 or more (S3: N
O), URA is calculated according to the above equation 6 (S5). Note that WstU in steps S4 and S5
RA represents the worst value of the URA Index. Next, σ _UARE is calculated using the URA calculated in step S4 or step S5 (S6), and the Mathematical Expression 1 is calculated using σ _UARE calculated in step S6.
2DRMS is calculated according to the definition of. 2 found here
DRMS is included in GPS output message
Is passed to.

Here, the use of 2DRMS in the application program executed by the host CPU of the car navigation system will be described with an example. For example, the car navigation system has a map matching function. In this case, the host application can use 2DRMS to determine which of the positioning solution by the GPS receiver 50 and the position based on the result of the map matching should be the final application output. That is, G
If the difference between the PS positioning solution and the position obtained by the map matching result is within 2DRMS, the difference is within the error range and the map matching result is adopted as it is.
If it exceeds 2DRMS, the positioning solution by the GPS receiver 50 is adopted because the map matching result is incorrect.

Here, 2DRMS by the GPS receiver 50
Since is an index that accurately reflects the actual positioning error as described above, it can be understood that when the result of map matching is erroneous, the effect of this error can be removed reliably and promptly. That is, by using the 2DRMS obtained by the calculation process of FIG. 2, the application program can improve the accuracy of the application output as the final vehicle position estimation result.

In the above embodiment, the 2DRMS calculation process (FIG. 2) is performed in the GPS receiver, and 2DRMS as the calculation result is output from the GPS receiver. However, 2DRMS calculation process (Fig. 2)
Can also be executed by the application program by receiving the necessary parameters from the GPS receiver.

[0036]

As described above, according to the present invention, G
URA can be further reflected in 2D RMS that takes EHPE into consideration as an error estimation value for the PS positioning solution. This makes it possible to adaptively estimate the error and accurately reflect the error in the 2DRMS even if the space segment and the control segment of the GPS system undergo fluctuations that improve or deteriorate the reliability.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a GPS receiver as an embodiment of the present invention.

FIG. 2 is a flowchart showing a 2DMRS calculation process executed by the GPS receiver of FIG.

FIG. 3 is a diagram for explaining three segments that compose the GPS system.

[Explanation of symbols]

1 antenna 2 Frequency converter 5 AD converter 6 Carrier correlation section 7 Carrier NCO 8 CA code correlation section 9 CA code generator 10 code NCO 10 Digital signal processor 11 CPU 12 memories 30 Digital signal processor 50 GPS receiver

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 2F029 AA02 AB07                 5J062 AA03 CC07 DD05 DD11 DD23                       DD24 DD25

Claims (7)

[Claims] 1. A HDOP (Horizontal) calculated using a Kalman filter for calculating a GPS positioning solution in a positioning calculation process.
DOP) and a horizontal (latitude, longitude) direction estimation error σ _{H_Kalman} calculated from the diagonal elements of the error covariance matrix obtained in the process of calculating the estimated value of the Kalman filter. 2DRMS (Distance Root Mean Square) as the accuracy evaluation value of the GPS positioning solution is calculated using the above-mentioned indexes by the following formula 1
And a GPS receiver that calculates and outputs according to the definition of Equation 2. [Equation 1] However, Where URA is the 1σ value of the error based on the URA Index σ _UserSeg is the 1σ value of the error for the user segment 2. The URA, where URA Index is N, N <
When 6, it is calculated by the definition of URA = 2 ^{1 + N / 2} , and N ≧
The GPS receiver according to claim 1, wherein the GPS receiver is calculated according to the definition of URA = 2 ^N−2 when 6. 3. The value N as a URA index used in the calculation of the URA is the worst value in the URA index for each GPS satellite receiving a GPS signal. GPS receiver. 4. The GPS receiver according to claim 1, further comprising: estimating the vehicle position using 2DRMS output by the GPS receiver as an accuracy evaluation value for a GPS positioning solution. Characteristic car navigation system. 5. HDOP, which receives a GPS signal to calculate a GPS positioning solution, uses a Kalman filter to calculate the GPS positioning solution, and is calculated in a positioning calculation process, and
Horizontal (latitude, longitude) calculated from the diagonal elements of the error covariance matrix obtained in the process of calculating the estimated value of the Kalman filter
In an apparatus capable of outputting each index of the direction estimation error σ _{H_Kalman} , 2DRMS as an accuracy evaluation value of a GPS positioning solution is calculated by using each of the indexes, using Equation 1 (Equation 1) and Equation 2 (Equation 2).
A method of calculating an accuracy evaluation value for a GPS positioning solution, characterized by: 6. The URA is calculated with the definition of URA = 2 ^{1 + N / 2} when N <6, and the definition of URA = 2 ^N-2 when N ≧ 6, where N <6. The method of calculating the accuracy evaluation value for the GPS positioning solution according to claim 5, wherein: 7. The value N as a URA index used in the calculation of the URA is the worst value in the URA index for each GPS satellite receiving a GPS signal. A method of calculating an accuracy evaluation value for the GPS positioning solution.