JP2005077318A - Global positioning system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a global positioning system for accurately detecting a satellite, where multipath, or the like occurred. <P>SOLUTION: A receiver position is calculated by master positioning arithmetic processing, based on all received GPS satellites, a receiver position is calculated by sub positioning arithmetic processing for three GPS satellites successively excluding a GPS satellite from respective GPS satellites one by one, a recursive residual APR is calculated, based on pseudo distance and approximation distance, based on all GPS satellites, the maximum difference of the processing result in the sub positioning arithmetic processing is calculated, and the presence or absence of the multipath is determined, based on the recursive residual APR and the maximum difference. When the multipath is occurring, at least two weighing operations are made to each GPS satellite successively assuming that each receiver position calculated by the sub positioning arithmetic operation is a true value, a position error is calculated, based on the operation result, and a satellite is determined to have generated a multipath when the position error increases proportionally to the increase of weight to the weight of only the GPS satellite excluded by the operation of the receiver position assumed to be a true value. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a global positioning system that measures a receiver position based on received radio waves from a plurality of GPS (Global Positioning System) satellites orbiting the earth.

  As a conventional global positioning system, for example, using the predicted position coordinates of the receiver obtained from the previous positioning result and the satellite position coordinates calculated using the orbit information from the satellite, the distance (approximate from the predicted position to the satellite) Distance) and an approximate distance calculator that calculates the propagation distance (pseudo distance) from the satellite to the receiver using the time difference between the satellite transmission time and the receiver reception time. If the difference between the pseudorange and approximate distance exceeds the predetermined allowable error range, it is determined that an error has occurred due to multipath, etc., and the pseudorange and approximate distance of the satellite are used for positioning calculation. By doing so, a navigation satellite signal receiver has been proposed in which the influence of error satellites such as multipath is eliminated (see, for example, Patent Document 1).

However, in the conventional example described in Patent Document 1, it is difficult to properly set a threshold for determining a multipath, and in particular, when the predicted position is not correct, the difference between the approximate distance and the pseudo distance Therefore, it is difficult to determine the occurrence of multipath.
In order to solve this problem, the following procedure has been proposed as a detection / isolation procedure for a satellite in which a multipath has occurred (for example, see Non-Patent Document 1).

(1) The pseudorange residual r is calculated using all received satellites (All-in-view Solution).
(2) If the pseudorange residual r is smaller than the threshold value RD, it is determined that all the receiving satellites are normal and the normality (Integrity) check is successful. On the other hand, if the pseudorange residual r is larger than the threshold value RD, it is determined that the normal check fails. To do.

(3) When failure of normal check is detected, respective pseudorange residuals r1, r2,... Rn are calculated for n subsets of n-1 satellites (Sub-set-solution).
(4) If the pseudorange residual ri is smaller than the threshold RI and the other pseudorange residuals r are larger than the threshold RI, the i-th satellite is a failed satellite.
If two or more pseudorange residuals ri are smaller than the threshold RI, a failed satellite cannot be detected.

(5) If the failed satellites can be separated, use the Sub-set-solution that excludes the failed satellites for positioning. On the other hand, if a failed satellite cannot be identified even if a normal check failure is detected, the All-in-view Solution is used for positioning. However, the position accuracy decreases.
Japanese Patent Laying-Open No. 2003-57327 (first page, FIG. 1) "GLOBAL POSITIONING SYSTEM, VOLUME V" The Institute of Navigation, Washington, DC, Journal of The Institute of Navigation Vol. 35, No. 2, Summer 1988 (see pages 49-65, especially page 54)

  In the conventional example described in Non-Patent Document 1, when it is determined that the normal check has failed, the satellites are sequentially removed one by one from the n satellites, and n−1 consisting of n−1 satellites. If the pseudorange residuals r1, r2,... Rn are calculated for a subset, and one pseudorange residual ri is smaller than the threshold RI and the other pseudorange residual r is larger than the threshold RI, the i-th satellite Although it can be identified as a failed satellite in which a multipath or the like has occurred, if two or more pseudorange residuals ri are smaller than the threshold RI, the failed satellite cannot be detected, There is an unresolved problem that the accuracy of detecting the failed satellite that has occurred is low.

  Accordingly, the present invention has been made paying attention to the unsolved problems of the above-described conventional example, and aims to provide a global positioning system capable of accurately detecting a satellite in which a multipath or the like has occurred. Yes.

  The first technical means includes an antenna unit for receiving radio waves from satellites, a detection unit for demodulating received signals from a plurality of satellites, and data analysis for collecting and analyzing satellite navigation messages from signals demodulated by the detection units. A transmission time conversion unit that calculates a transmission time at which a signal is transmitted from the satellite, and a pseudo-range that calculates a pseudo distance that represents a propagation distance of radio waves from a difference between the transmission time at the satellite and the reception time at the receiver. In a global positioning system including a distance calculation unit, and a positioning calculation unit that calculates a receiver position based on the pseudo distance calculated by the pseudo distance calculation unit and the satellite coordinate position, the positioning calculation unit receives a received signal Master positioning calculation means for calculating the receiver position based on the pseudoranges and satellite positions of all the satellites that received the satellite, and the pseudoranges and satellite positions of the remaining satellites except for one of all the satellites in order. Based on compound Sub positioning calculation means for calculating N receiver positions, an approximate distance calculation section for calculating an approximate distance representing a coordinate space distance from the satellite coordinate position to the receiver approximate coordinate position, and calculation by the master positioning calculation means Inductive residual calculation means for calculating an inductive residual obtained by adding a square value of a deviation between the approximate distance calculated for each satellite calculated by the approximate distance calculation unit and the pseudo distance based on the received receiver position and satellite coordinates A maximum difference calculation means for calculating a maximum difference between a plurality of receiver positions calculated by the sub-positioning calculation means, and a multipath determination means for determining the presence or absence of multipath generation based on the recursive residual and the maximum difference And when the determination result of the multipath determination means indicates that there is no multipath, the normal time selection means for selecting the receiver position calculated by the master positioning calculation means, and the determination result of the multipath determination means. When multipath occurs, weighting calculation is performed for each satellite, the satellite where the multipath is generated is identified from the calculation result, and the reception received by the sub-positioning calculation means excluding the satellite where the multibus is generated Multipath generation state selection means for selecting the machine position is provided.

  In the first technical means, the master positioning calculation means calculates the receiver position based on the pseudoranges and the satellite positions of all the satellites that have received the received signals, and the sub-positioning calculation means sequentially from all the satellites. The receiver position is calculated based on the pseudorange and satellite position of the remaining satellites except for one, and the multipath determination means adds the square value of the deviation between the pseudorange and approximate distance for each satellite. The presence or absence of multipath is determined based on the residual and the maximum difference between the receiver positions calculated by the sub-positioning calculation means. If no multipath occurs, the receiver position calculated by the master positioning calculation means is When a multipath occurs, a weighting operation is performed for each satellite, the satellite where the multipath has occurred is identified from the calculation result, and the satellite from which the multipath has occurred is excluded. Selecting a receiver position computed by the positioning calculation means as a positioning result.

For this reason, even if it is not possible to identify the satellite where the multipath occurred from only the calculation result of the sub-positioning calculation means, the multipath occurred by performing the weighting calculation and identifying the satellite where the multipath occurred from the calculation result. The satellite can be accurately identified and the receiver position can be accurately measured.
Further, the second technical means is the first technical means, wherein the multipath determining means is a multipath in which the recursive residual is less than a recursive residual threshold and a maximum deviation is less than a maximum deviation threshold. The multi-path non-occurrence state is determined when the path non-occurrence condition is satisfied, and the multi-path occurrence state is determined when the multi-path non-occurrence condition is not satisfied.

  In the second technical means, the recursive residual APR, which is a sum of square values obtained by squaring the pseudo-range and approximate distance deviation of each satellite in the multipath determination means, is less than the recursive residual threshold TH1, In addition, it is determined whether or not the multipath non-occurrence condition in which the maximum deviation MS of the receiver position calculated by the sub-positioning calculation means is less than the maximum deviation threshold TH2 is satisfied, and when APR <TH1 and MS <TH2, Since it is determined that a non-occurrence state has occurred, and when APR ≧ TH1 and / or MS ≧ TH2, it is determined that a multi-path occurrence state has occurred, it is possible to accurately determine whether a multi-path has occurred.

  Further, the third technical means is the first or second technical means, wherein the multipath occurrence state selecting means sequentially sets a plurality of N receiver positions calculated by the sub-positioning calculating means as true values in order, for each satellite. A position error calculation means for calculating a position error for each of at least three positioning values obtained by weighting at least two different positioning values and a positioning value calculated by the master positioning calculation means, and the position error calculation means Among the position errors, when the position error increases in proportion to the increase of the weight for the satellite that is excluded from the calculation of the receiver position set to the true value, the satellite is identified as the satellite that generated the multipath. It is characterized by being configured.

  In this third technical means, a plurality of N receiver positions are successively set as true values, and at least two differently weighted positioning values for each satellite and three positioning values calculated by the master positioning calculating means are used. When the position error increases in proportion to the increase in weight for only the satellites that are excluded from the calculation of the receiver position set to the true value among the calculated position errors. By identifying the satellite as a satellite in which a multipath has occurred, the satellite in which the multipath has occurred can be accurately detected.

  Furthermore, the fourth technical means is that the multipath generation state selection means is for the satellites excluded from the calculation of the receiver position set to a true value among the position errors calculated by the position error calculation means. When there are multiple pairs in which the position error increases in proportion to the increase of the weight, the satellite having the largest increase rate of the position error with respect to the increase of the weight is specified as the satellite in which the multipath has occurred. It is characterized by being.

  In the fourth technical means, the position error increases in proportion to the increase in weight with respect to the satellites that are removed by the calculation of the set receiver position among the position errors calculated by the position error calculation means. When there are a plurality of satellites, the satellite having the largest increase rate of the position error relative to the increase in weight is specified as the satellite in which the multipath has occurred, so that the satellite in which the multipath has occurred can be accurately detected.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing an embodiment of the present invention. In the figure, 1 is a global positioning system, and at least four GPS (Global Positioning System) satellites SV14 and SV20 orbiting outside the earth's atmosphere. The radio wave transmitted from the SV 22 is received by the GPS receiver 2 to obtain the receiver position.

The GPS satellites SV14, SV20 to SV22 are spread spectrum communication using a direct spread method for a receiver, two signals of 157.64 MHz in the L-band frequency band and 1227.6 MHz for correcting the propagation delay of the ionosphere. Send by method.
The GPS receiver 2 includes an antenna 3 that receives radio waves from GPS satellites SV14 and SV20 to SV22, a low noise amplifier 4 that amplifies an RF signal received by the antenna 3, and an RF amplified by the low noise amplifier 4. The frequency conversion unit 5 down-converts the signal into an IF signal, and the GPS demodulation / calculation unit 6 to which the IF signal converted by the frequency conversion unit 5 is input.

The GPS demodulator / arithmetic unit 6 demodulates the IF signal received from each of the GPS satellites SV14, SV20 to SV22 output from the frequency converter 5, and generates digital data including time information, orbit information, and the like. Is used to calculate the positioning data representing the receiver position.
The GPS demodulation / calculation unit 6 executes positioning calculation processing. Here, as shown in FIG. 2 in advance, the relationship between the satellite number and the variable i is as follows: the satellite SV21 variable is “1”, the satellite SV14 variable is “2”, the satellite SV20 variable is “3”, and the satellite SV22 Set the variable to be “4”.

As shown in FIG. 3, in the positioning calculation process, first, in step S1, each GPS satellite SV14, SV20, SV21, SV22 transmits a transmission time and a reception time when the GPS receiver 2 receives a radio wave. The pseudo distance ym _i (i = 1, 2,... 4) = c × t _i corresponding to the propagation distance of the radio wave is calculated from the difference t _i and the speed of light c.
Next, the process proceeds to step S2, and the satellite position (X _i , Y _i , Z _i ) is calculated using the satellite transmission time and orbit information included in the demodulated digital data.

Then, it goes to step S3, the receiver position and (x, y, z), when the deviation between the reference time and the receiver time of the GPS satellites and the Delta] t, the four unknowns, pseudorange ym _i And the master positioning calculation is performed by solving the following equations (1) to (4) based on the satellite position (X _i , Y _i , Z _i ).
ym ₁ = {(X ₁ −x) ² + (Y ₁ −y) ² + (Z ₁ −z) ² } ^1/2 + cΔt (1)
ym ₂ = {(X ₂ −x) ² + (Y ₂ −y) ² + (Z ₂ −z) ² } ^1/2 + cΔt (2)
ym ₃ = {(X ₃ −x) ² + (Y ₃ −y) ² + (Z ₃ −z) ² } ^1/2 + cΔt (3)
ym ₄ = {(X ₄ −x) ² + (Y ₄ −y) ² + (Z ₄ −z) ² } ^1/2 + cΔt (4)
Next, the process proceeds to step S4, and the approximate distance yp between the first to fourth satellites from the calculated receiver position (x, y, z) and satellite position (X _i , Y _i , Z _i ). It is calculated based on the _{1 ~yp 4} below (5) to (8).

yp ₁ = {(X ₁ -x) ² + (Y ₁ -y) ² + (Z ₁ -z) ² } ^1/2 (5)
yp ₂ = {(X ₂ −x) ² + (Y ₂ −y) ² + (Z ₂ −z) ² } ^1/2 (6)
yp ₃ = {(X ₃ −x) ² + (Y ₃ −y) ² + (Z ₃ −z) ² } ^1/2 (7)
yp ₄ = {(X ₄ -x) ² + (Y ₄ -y) ² + (Z ₄ -z) ² } ^1/2 (8)
Next, the process proceeds to step S5, and the recursive residual APR is calculated by performing the following equation (9) based on the pseudo distance ym _i and the approximate distance yp _i .

次いで、ステップＳ６に移行して、時刻ずれΔｔが求まっているので、前記（１）〜（４）式から例えば（１）式から順に（２）式、（３）式及び（４）式を除いた３つの式からＮ個即ち４個の受信機位置（ｘ_i ，ｙ_i ，ｚ_i ）を演算するサブ測位演算を行う。

Next, the process proceeds to step S6, and the time lag Δt is obtained. Therefore, from the equations (1) to (4), for example, the equations (2), (3), and (4) are sequentially calculated from the equation (1). Sub-positioning calculation is performed to calculate N receiver positions (x _i , y _i , z _i ) from the three expressions excluded.

Next, the process proceeds to step S7, the maximum difference MS (Maximum Separation) of the sub-positioning result calculated in step S6 is calculated, and then the process proceeds to step S8.
In this step S8, the multipath non-occurrence condition in which the recursive residual APR calculated in step S5 is less than a preset recursive residual threshold TH1 and the maximum difference MS is less than a preset maximum differential threshold TH2 is set. If APR <TH1 and MS <TH2 and the multipath non-occurrence condition is satisfied, the process proceeds to step S9 and uses the four satellites calculated in step S3. The receiver position (x _M , y _M , z _M ) calculated in the master positioning calculation is selected as the positioning result, and the process returns to step S1.

On the other hand, if the determination result in step S8 is APR ≧ TH1 or MS ≧ TH2, or APR ≧ TH1 and MS ≧ TH2, the multipath non-occurrence condition is not satisfied, and it is determined that the multipath is in the state. The process proceeds to S10.
In step S10, the weighting algorithm shown in FIG. 4 for identifying the GPS satellite in which the multipath has occurred is executed, and then the process returns to step S1.

As shown in FIG. 4, this weighting algorithm is a sub-position calculation result obtained by first setting the variable i to “1” in step S11 and then moving to step S12 to exclude the GPS satellite of the variable i. After setting the receiver position (x _i , y _i , z _i ) as a true value (x ^* _i , y ^* _i , z ^* _i ), the process proceeds to step S13.
In step S13, the weight w _{i is set} to 2 ⁻¹ for the GPS satellite of the variable i. 2. Weighting calculation processing when ⁺¹ is performed.
This weighting process applies the Cancel-B method as a normal GPS positioning calculation model,
y _i = H _i X + b (10)
y _i −y ^* = (H _i −H ^* ) X (11)
Here, y _i is a difference vector between the pseudo distance and the approximate distance, H is a gaze direction vector, and X is a position error vector. However,

これに対し下記のように重み付け法を適用する。
ｗ_i （ｙ_i −ｙ^* ）＝ｗ_i （Ｈ_i −Ｈ^* ）Ｘ …………（１２）
ただし、
ｙ^* ＝（Σｗ_i ｙ_i ）／（Σｗ_i ）
Ｈ^* ＝（Σｗ_i Ｈ_i ）／（Σｗ_i ）
そして、重み付け法を適用した変数iのGPS衛星に対して、重み毎の受信機位置（ｘ_i1，ｙ_i1，ｚ_i1），（ｘ_i2，ｙ_i2，ｚ_i2） を算出する。

On the other hand, the weighting method is applied as follows.
w _i (y _i −y ^* ) = w _i (H _i −H ^* ) X (12)
However,
y ^* = (Σw _i y _i ) / (Σw _i )
H ^* = (Σw _i H _i ) / (Σw _i )
Then, the receiver position (x _i1 , y _i1 , z _i1 ) and (x _i2 , y _i2 , z _i2 ) for each weight is calculated for the GPS satellite of the variable i to which the weighting method is applied.

Next, the process proceeds to step S14, and the following formula is calculated based on the receiver position for each weight calculated in step S13 and the true values (x ^* _i , y ^* _i , z ^* _i ) set in step S12. To calculate a position error PE and store it in a storage device such as a memory.
PE = {(x _ij −x ^* ) ² + (y _ij −y ^* ) ² + (z _ij −z ^* ) ² } ^1/2 (13)
Next, the process proceeds to step S15, the variable i is incremented (i = i + 1), and then the process proceeds to step S16 to determine whether or not the variable i exceeds the number N of received satellites, and i ≦ N. Sometimes it is determined that the position error calculation has not ended, and the process returns to step S12. When i> N, it is determined that the position error calculation has ended, and the process proceeds to step S17.

  In this step S17, the position error PE with respect to the weight for each GPS satellite stored in the storage device is read, and the position error PE increases in proportion to the increase of the weight for only the satellites excluded from the sub-operation. For the remaining three satellites, a set satisfying the multipath generation condition in which the position error PE decreases in inverse proportion to the increase in weight is extracted, and then the process proceeds to step S18 to satisfy the extracted multipath generation condition. It is determined whether or not there is only one set. If there is only one set, the process proceeds to step S19, and the GPS satellites excluded from the calculation of the receiver position set as the true value of the set are multipathed. Then, the process proceeds to step S20, where the receiver positions calculated by the three GPS satellites excluding the corresponding multipath generating satellite in the sub-positioning process are set as positioning results. Then, the weighting algorithm is terminated and the process returns to step S1 in FIG.

Further, when there are a plurality of sets whose determination result of step S18 satisfies the multipath generation condition, the process proceeds to step S21, and the set with the maximum change rate of the position error PE that increases due to the increase of the weight is obtained. After the GPS satellite excluded from the calculation of the receiver position set as the true value is specified as the multipath generation satellite, the process proceeds to step S20.
The processing in FIG. 3 corresponds to the positioning calculation means, of which the processing in step S1 corresponds to the pseudo distance calculation means, the processing in steps S2 to S3 corresponds to the master positioning calculation means, and the processing in step S5 is functional. Corresponding to the residual calculation means, the processing in step S6 corresponds to the sub-positioning calculation means, the processing in step S7 corresponds to the maximum difference calculation means, the processing in step S8 corresponds to the multipath determination means, and in step S9 The process corresponds to a normal selection unit, and the process of step S10 and the process of FIG. 4 correspond to a multipath generation state selection unit.

Next, the operation of the above embodiment will be described.
Assuming that the GPS satellite in which the multipath is generated is, for example, the GPS satellite SV21 and is known and the true value (x *, y *, z *) of the receiver is known, each GPS satellite SV21, SV14, SV20, when the position error PE of SV22 computed by applying the equation (13), as shown in FIG. 5, the GPS satellites SV21 is multipath generation satellites, the weight is 2 ⁰ state gentle as in the position error PE is gradually increased as the weight from the state of the order of 170m is increased and 2 ^{1, 2} 2, 2 ^3, weight conversely 2 ^-1, 2 ^-2, decreases with 2 ^-3 Until the position error becomes substantially “0”, and thereafter the position error remains substantially “0” even when the weight is reduced.

However, for the remaining three GPS satellites SV14, SV20, and SV22 that do not generate multipath, the position error PE decreases relatively steeply when the weight increases to 2 ⁻³ , 2 ⁻² , 2 ⁻¹ , 2 ^0. Thereafter, when the weight increases from 2 ⁰ to 2 ¹ , 2 ² , 2 ³ , the position error PE decreases relatively slowly.
From this result, the position error PE increases in proportion to an increase in weight for a GPS satellite in which a multipath has occurred, whereas it increases in inverse proportion to an increase in weight for a GPS satellite in which no multipath has occurred. It is understood that the position error PE is reduced.

  Based on this principle, in the above-described embodiment, if the GPS receiver 2 receives radio waves from the four GPS satellites SV21, SV14, SV20, SV22 as described above, the GPS receiver 2 receives the signals via the antenna 3. The RF signal is amplified by the low noise amplifier 4 and then down-converted to an IF signal by the frequency converter 5, and this IF signal is supplied to the GPS demodulator / arithmetic unit 6 to demodulate the IF signal and transmit time, Digital data including orbit information is generated.

Then, the processing of FIGS. 3 and 4 is executed based on the demodulated digital data of each GPS satellite to determine the presence / absence of a GPS satellite in which a multipath has occurred, and the receiver position is calculated based on the determination result.
That is, as described above, the pseudo distances ym ₁ , ym ₂ , ym ₃ , ym ₄ are calculated for all the received GPS satellites SV 21, SV 14, SV 20, SV 22.

Then, from the calculated pseudo distance ym _i and the satellite position (X _i , Y _i , Z _i ), the equations (1) to (4) are solved to obtain the receiver position (x _M , y _M , z _M ). Performs the master positioning calculation to calculate.
Next, based on the receiver position (x _M , y _M , z _M ) obtained by the master positioning calculation process and the satellite positions (X _i , Y _i , Z _i ) of each GPS satellite, the above formulas (5) to (5) 8) The approximate distance ym _i for each GPS satellite is calculated by calculating the equation.

Then, an inductive residual APR is calculated based on the pseudo distance ym _i and the approximate distance yp _i .
Furthermore, sub-positioning calculation processing is performed based on the remaining three GPS satellites, each of which is excluded from four GPS satellites, and the receiver position that is the four sub-positioning solutions is calculated, and further calculated by sub-positioning calculation processing. The maximum difference MS of the received receiver position is calculated.
Then, it is determined whether or not the calculated recursive residual APR satisfies a multipath non-occurrence condition in which the recursive residual threshold APR is less than the recursive residual threshold TH1 and the maximum difference MS is less than the maximum difference threshold TH2. When the generation condition is satisfied, it is determined that no multipath has occurred in each of the GPS satellites SV21, SV14, SV20, and SC22, and the receiver position (x _M , y _M , z _M calculated by the master positioning calculation process is determined. ) Is selected as a true value, and this is output to a subsequent car navigation system, for example.

However, when a multipath occurs in the GPS satellite SV21 and the calculated recursive residual APR and maximum difference MS do not satisfy the multipath non-occurrence condition, the weighting algorithm of FIG. 4 is executed.
In this weighting algorithm, first, the variable i is set to “1”, and sub-positioning calculation processing is performed based on three GPS satellites SV14, SV20, SV22 excluding the GPS satellite SV21 among the four GPS satellites. The received receiver position (x _1, y ₁ , z ₁ ) is a true value (x ^* _, y ^* , Z ^* ) And a weighting operation in which two weights represented by 2 ⁻¹ and 2 ⁺¹ are set for each satellite, and receiver positions (x ₁₁ , y ₁₁ , z ₁₁ ), (x ₁₂ , y ₁₂ , z ₁₂ ) to (x ₄₁ , y ₄₁ , z ₄₁ ), (x ₄₂ , y ₄₂ , z ₄₂ ) are calculated, and based on these, the calculation of the formula (13) is performed, and the position for each weight is calculated. The error PE is calculated. As shown in FIG. 6, when the multipath generating satellite is the GPS satellite SV21, the calculation result indicates that the receiver position relative to the GPS satellite SV21 excluded from the calculation of the receiver position assumed to be a true value is shown in FIG. The error PE increases in proportion to the increase in weight as indicated by a circle, and for the remaining GPS satellites SV14, SV20, SV22, the receiver position error PE is indicated by □, ▲, and ◆, It tends to decrease in inverse proportion to the increase in weight.

Next, the variable i is set to “2” and the receiver position (x _2, y ₂ , z ₂ ) calculated by performing the sub-positioning calculation based on the three GPS satellites SV21, SV20, SV22 excluding the GPS satellite SV14 is true. Value (x ^* _, y ^* , Z ^* ) And a weighting operation in which weights of 2 ⁻¹ and 2 ⁺¹ are set for each satellite, and receiver positions (x ₁₁ , y ₁₁ , z ₁₁ ), (x ₁₂ , y ₁₂ , z ₁₂ ) are assumed. ) To (x ₄₁ , y ₄₁ , z ₄₁ ), (x ₄₂ , y ₄₂ , z ₄₂ ), and based on these, the calculation of the equation (13) is performed to calculate the position error PE for each weight. . As shown in FIG. 7, the calculation result shows that the position error PE of the receiver increases in proportion to the increase in weight with respect to the GPS satellites SV14 and SV22 indicated by □ and ♦, and is indicated by ◯ and ▲. For the GPS satellites SV21 and SV20, the position error PE tends to decrease in inverse proportion to the increase in weight.

Furthermore, the variable i is set to “3”, and the receiver position (x _3, y ₃ , z ₃ ) calculated by performing the sub-positioning calculation based on the three GPS satellites SV21, SV14, SV22 excluding the GPS satellite SV20 is true. Value (x ^* _, y ^* , Z ^* ) And a weighting operation in which weights of 2 ⁻¹ and 2 ⁺¹ are set for each satellite, and receiver positions (x ₁₁ , y ₁₁ , z ₁₁ ), (x ₁₂ , y ₁₂ , z ₁₂ ) are assumed. ) To (x ₄₁ , y ₄₁ , z ₄₁ ), (x ₄₂ , y ₄₂ , z ₄₂ ) are calculated, and the calculation of the equation (13) is performed based on these to calculate the position error PE for each weight. . As shown in FIG. 8, the calculation result shows that the position error of the receiver increases in proportion to the increase in weight with respect to the GPS satellites SV20 and SV22 indicated by ▲ and ◆, and GPS indicated by ◯ and □. For satellites SV21 and SV14, the receiver position error tends to decrease in inverse proportion to the increase in weight.

Furthermore, the variable i is set to “4”, and the receiver position (x _4, y ₄ , z ₄ ) calculated by performing the sub-positioning calculation based on the three GPS satellites SV21, SV14, SV20 excluding the GPS satellite SV22. True value (x ^* _, y ^* , Z ^* ) And a weighting operation in which weights of 2 ⁻¹ and 2 ⁺¹ are set for each satellite, and receiver positions (x ₁₁ , y ₁₁ , z ₁₁ ), (x ₁₂ , y ₁₂ , z ₁₂ ) To (x ₄₁ , y ₄₁ , z ₄₁ ), (x ₄₂ , y ₄₂ , z ₄₂ ) are calculated, and based on these, the calculation of the equation (13) is performed to calculate the position error PE for each weight. . As shown in FIG. 9, the calculation result shows that the position error PE increases in proportion to the increase of the weight with respect to the GPS satellites SV14, SV20, and SV22 indicated by □, ▲, and ◆, and GPS indicated by ◯ For the satellite SV21, the position error PE tends to decrease in inverse proportion to the increase in weight.

Therefore, each GPS satellite SV21, SV14. Assuming that the receiver position calculated in the sub-positioning calculation process in which SV20 and SV22 are sequentially removed is a true value, the weighting calculation is performed on the basis of the receiver position and the true value of each GPS satellite. As understood from FIGS. 6 to 9 showing the calculation results by performing the calculation, only the weight of the GPS satellite removed from the calculation of the receiver position of the sub-positioning calculation process in which only FIG. 5 is set as the true value. Thus, the position error increases in proportion to the increase in weight, and the multipath generation condition in which the position error decreases in inverse proportion to the increase in weight with respect to the weights of the remaining GPS satellites is satisfied. The GPS satellite SV21 is specified as the multipath generating satellite, the receiver position (x _1, y ₁ , z ₁ ) assumed as the true value is determined as the positioning result, and this is output to the car navigation system or the like at the subsequent stage.

In addition, when there are a plurality of results satisfying the multipath generation condition, a GPS satellite that has selected the largest change rate of the position error PE with respect to the increase of the weight and excluded the true value of the set is calculated. Is determined as a multipath generating satellite, the receiver position (x _i, y _i , z _i ) assumed as the true value of this set is determined as a true value, and this is output to the car navigation system or the like at the subsequent stage.

Thus, according to the above embodiment, the recursive residual APR calculated based on the pseudo distance ym _i calculated by the master positioning calculation process and the approximate distance yp _i and one of all received GPS satellites in order. The presence or absence of a multipath-occurring GPS satellite is determined based on the maximum difference MS of the receiver position calculated by performing sub-positioning calculation processing on the remaining GPS satellites, and there is no GPS satellite in which a multipath has occurred In this case, the receiver position calculated in the master positioning calculation process is selected as a true value, and if there is a GPS satellite generated by a multipath satellite, each receiver position calculated in the sub-positioning calculation process is sequentially true. Assuming the value, weighting processing is performed for the four GPS satellites at this time, the position error PE is calculated for them, and the calculation result is assumed to be a true value. The position error increases in proportion to the increase in the weight with respect to the weight of the GPS satellite excluded by the calculation of the receiver position, and the position error in inverse proportion to the increase in the weight for the remaining three GPS satellites. When the reduced multipath generation condition is satisfied, the excluded GPS satellite is specified as the multipath generation satellite. Therefore, when the presence of the multipath generation satellite is detected, the multipath generation satellite can be accurately specified. .

  The recursive residual APR is calculated with the master positioning solution that uses all the received satellites, and the recursive residual APR is calculated with the sub-positioning solution that uses three satellites that exclude the GPS satellites one by one. Then, as shown in Table 1 below, the APR is 149.8 in the master positioning solution, and all the recursive residuals APR are 0 in the remaining sub-positioning solutions. The position error is 163.9 m in the master positioning solution, 18.2 m in the sub-positioning solution 21 using three satellites excluding the GPS satellite SV21, and 1993.0 m in the sub-positioning solution 14 using three satellites excluding the GPS satellite SV14. With the three satellites excluding the GPS satellite SV20, the sub-positioning solution 20 is 962.7 m, and with the three satellites excluding the GPS satellite SV22, the sub-positioning solution 22 is 221.2 m. In the described conventional example, since there are a plurality of sub-positioning solutions that are “0”, the master positioning solution is selected, so that a position error of 163.9 m occurs. In the present embodiment, As described above, when the sub-positioning solution 21 is selected, the position error becomes the minimum value 18.2 and the more accurate receiver position can be obtained. It can be calculated to.

なお、上記実施形態においては、重み付けアルゴリズムとして通常のＧＰＳ測位計算モデルとしてCancel-B法を適用した場合について説明したが、これに限定されるものではなく、通常のＧＰＳ測位計算モデルを下記のように設定して重み付け演算を行うようにしてもよい。

In the above embodiment, the case where the Cancel-B method is applied as a normal GPS positioning calculation model as a weighting algorithm has been described. However, the present invention is not limited to this, and a normal GPS positioning calculation model is as follows. May be set to perform weighting calculation.

Y = H · X
Here, Y is a difference vector between the pseudo distance and the approximate distance, H is a gaze direction vector, and X is a position error vector.
Then, it is expected that the position error can be reduced by assigning an appropriate weight according to the measurement error at the time of measuring each satellite.
If weight vector w = (w ₁ , w _2, ... W _n ),
w ・ Y = w ・ H ・ X
(WH) ^T wY = (wH) ^T wHX
H ^T WY = H ^T WHX
X = (H ^T WH) ⁻¹ H ^T WY
However,

また、上記実施形態では４つの衛星からの電波を受信可能な場合について説明したが、これに限定されるものではなく、５以上の衛星からの電波を受信可能な場合でも、上記実施形態と同様にマルチパス衛星を特定することができる。

In the above embodiment, the case where radio waves from four satellites can be received has been described. However, the present invention is not limited to this, and even when radio waves from five or more satellites can be received, the same as in the above embodiment. Multipath satellites can be identified.

It is a block diagram which shows one Embodiment of this invention. It is a figure which shows the table showing the relationship between the received satellite number and a variable. It is a flowchart which shows a positioning calculation process. It is a flowchart which shows the weighting calculation process of FIG. It is a characteristic diagram which shows the relationship between the weight for every GPS satellite, and a position error. It is a characteristic diagram when the sub-positioning solution excluding the GPS satellite SV21 is a true value. It is a characteristic diagram when the sub-positioning solution excluding the GPS satellite SV14 is a true value. It is a characteristic diagram when the sub-positioning solution excluding the GPS satellite SV20 is a true value. It is a characteristic diagram when the sub-positioning solution excluding the GPS satellite SV22 is a true value.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Global positioning system, 2 ... GPS receiver, SV21, SV14, SV20, SV22 ... GPS satellite, 3 ... Antenna, 4 ... Low noise amplifier, 5 ... Frequency conversion part, 6 ... GPS demodulation and calculation part

Claims (4)

An antenna unit that receives radio waves from a satellite, a detection unit that demodulates received signals from a plurality of N satellites, a data analysis unit that collects and analyzes satellite navigation messages from signals demodulated by the detection unit, and a satellite A transmission time conversion unit for calculating a transmission time at which a signal is transmitted, a pseudo distance calculation unit for calculating a pseudo distance representing a propagation distance of radio waves from a difference between the transmission time at the satellite and the reception time at the receiver, In a global positioning system including a positioning calculation unit that calculates a receiver position based on a pseudorange calculated by the pseudorange calculation unit and a satellite coordinate position,
The positioning calculation unit includes master positioning calculation means for calculating the receiver position based on the pseudoranges and satellite positions of all the satellites that have received the received signals, and the remaining satellites except for one of all the satellites in sequence. Sub-position calculating means for calculating a plurality of N receiver positions based on the pseudo-range and the satellite position, and an approximate distance for calculating an approximate distance representing a coordinate space distance from the satellite coordinate position to the receiver approximate coordinate position Based on the receiver position and satellite position calculated by the calculation unit and the master positioning calculation means, an inductive value obtained by adding the square value of the deviation between the approximate distance calculated by the approximate distance calculation unit and the pseudo distance Inductive residual calculation means for calculating a residual, maximum difference calculation means for calculating a maximum difference in receiver position calculated by the sub-positioning calculation means, and multipath generation based on the inductive residual and the maximum difference Whether or not Multipath determination means, a normal time selection means for selecting a receiver position calculated by the master positioning calculation means when the determination result of the multipath determination means is no multipath, and determination by the multipath determination means When there is a multipath occurrence, weighting calculation is performed for each satellite, the satellite where the multipath occurred is identified from the calculation result, and the sub-positioning calculation means excluding the satellite where the multibus is generated is used. A global positioning system comprising multipath generation state selection means for selecting a receiver position.   The multipath determination means is in a multipath non-occurrence state when the recursive residual is less than a recursive residual threshold and satisfies a multipath non-occurrence condition in which a maximum deviation is less than a maximum deviation threshold. The global positioning system according to claim 1, wherein the global positioning system is configured to determine and determine that the multipath generation state is satisfied when the multipath non-occurrence condition is not satisfied.   The multipath occurrence state selection means includes a plurality of N receiver positions calculated by the sub-positioning calculation means as true values sequentially, a positioning value obtained by performing at least two different weights for each satellite, and the master positioning calculation means. For each of the three positioning values calculated, the position error calculating means for calculating the position error and the position error calculated by the position error calculating means are excluded from the calculation of the receiver position set to the true value. 3. Only a selected satellite is configured to identify the satellite as a multipath-generated satellite when the position error increases in proportion to an increase in weight. The global positioning system described in.   The multipath generation state selection unit is configured to increase the weight of the satellite with respect to a satellite that is excluded from the calculation of the receiver position set to a true value among the position errors calculated by the position error calculation unit. The satellite is configured to identify a satellite having a maximum position error increase rate with respect to an increase in weight as a satellite in which a multipath has occurred when there are a plurality of proportionally increasing pairs. The global positioning system according to 3.

Images