JP2010071686A - Positioning apparatus, computer program, and positioning method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a positioning apparatus, a computer program, and a positioning method which can speed up ambiguity calculations and increase a period of time allowing the positioning of RTK in the urban area. <P>SOLUTION: In the positioning apparatus, a satellite communications section 12 includes a code phase measuring section 121 for measuring the pseudo-distance to each satellite, a carrier phase measuring section 122 for measuring the phase of the carrier of each satellite, and a carrier frequency measuring section 123 for measuring the frequency of the carrier of each satellite. A processing section 11 measures the frequency of the carrier of each satellite to determine a Doppler shift, thereby calculating the velocity vector of an object to be positioned 100. The processing section 11 determines the center B of a solution search range by B=A+T*V, with the position of the object to be positioned 100 before positioning being A, the velocity vector being V, and the time interval of velocity vector calculations being T and determines ambiguity with the solution search range limited. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a positioning device that measures the position of a positioning target such as a vehicle or a person using a signal transmitted from a GNSS positioning satellite (for example, a GPS positioning satellite), and a computer program for realizing the positioning device. And a positioning method.

  A signal transmitted from a positioning satellite is received by a receiving device (positioning target), a carrier phase (carrier phase) included in the received signal is measured, and the receiving device is based on the measured carrier phase and the position of the positioning satellite. Positioning methods for measuring the position of the are used in various fields.

  In the above positioning method, for example, a carrier phase difference (single phase difference) of a positioning satellite (other than the reference) with respect to a reference positioning satellite is obtained from a plurality of positioning satellites, and a single phase difference difference (two (Multiple phase difference). This phase difference can be divided into an integer part and a decimal part when converted into wave number units. The phase that can be measured by the receiving apparatus is a fractional part within a range of 2π (0 to 360 deg), and an integer part (integer multiple of 2π) cannot be directly measured. This integer part is referred to as ambiguity, integer bias, or the like.

  In a positioning device using the carrier phase as described above (for example, an RTK-GPS receiver), once the ambiguity can be calculated, the positioning is continuously performed as long as the signal transmitted by each positioning satellite is captured. It is possible to measure the position of the object. However, if the radio wave from the positioning satellite is interrupted once (for example, if the cycle slips), it is necessary to calculate the ambiguity again, and it takes several seconds to several tens of seconds before the positioning can be resumed. was there.

  For this reason, for example, in Patent Document 1, a new integer bias (ambiguity) is estimated when the number of positioning satellites that can receive radio waves increases or decreases, or when the reference positioning satellite is switched. In addition, there is disclosed a carrier phase relative positioning device that can reduce the calculation time required for positioning by estimating the floating ambiguity after the change using the baseline vector estimated before the change.

In Patent Document 2, the position of the positioning station obtained during positioning is held during non-positioning, the calculated phase difference is obtained from the held positioning position at the start of positioning, and the decimal part of the calculated phase difference is obtained. By correcting the calculated phase difference so that it agrees with the observed phase difference in the range of ± 0.5 cycles and using the corrected calculated phase difference as an integer bias, positioning can be started immediately after power-on. A positioning device is disclosed.
JP 2003-185728 A Japanese Patent Laid-Open No. 2003-194915

  However, although the device of Patent Document 1 shows a high-speed ambiguity estimation method when the number of satellites that can receive (observable) radio waves increases or decreases, most of the satellites in the tunnel or under the overhead are used. It was difficult to determine ambiguity at a high speed when it was not possible to receive radio waves from the satellite, and it was necessary to develop a new method to expand the utilization rate of high-precision positioning in urban areas. In particular, in urban areas where there are many shielding objects such as buildings, cycle slips occur frequently, so each time it takes a long time to calculate ambiguity and receive RTK (real-time kinematic) services. There was a problem that the time that could be shortened.

  Further, in the apparatus of Patent Document 2, since the calculation phase difference is corrected within a range of ± 0.5 cycle of the decimal part of the calculation phase difference, for example, a positioning target that moves at a relatively high speed like a vehicle is used. As described above, during a non-positioning, when one cycle of the carrier phase difference moves beyond about one half of the distance converted, the integer bias cannot be determined, and the time for positioning can be determined. Could not increase.

  The present invention has been made in view of such circumstances, and a positioning device capable of speeding up the calculation of ambiguity and greatly increasing the time during which RTK positioning is possible in an urban area or the like, and the positioning device An object of the present invention is to provide a computer program and a positioning method for realizing the above.

  A positioning device according to a first aspect of the present invention includes a receiving unit that receives each carrier transmitted by a plurality of positioning satellites, and performs positioning based on the phase of each carrier received by the receiving unit. Position acquisition means for acquiring its own position, speed vector calculation means for calculating a speed vector from the position acquired by the position acquisition means, position acquired by the position acquisition means, and speed calculated by the speed vector calculation means Ambiguity determining means for determining an ambiguity included in a phase using a vector, and positioning means for positioning its own position based on the ambiguity determined by the ambiguity determining means. Features.

  A positioning apparatus according to a second invention is characterized in that, in the first invention, the velocity vector calculating means is configured to calculate a velocity vector based on a Doppler shift of each carrier wave received by the receiving unit. To do.

  The positioning device according to a third aspect of the present invention comprises, in the first aspect of the present invention, speed information acquisition means for acquiring information related to speed and angle acquisition means for acquiring a turning angle, wherein the speed vector calculation means is the speed information acquisition means. The speed vector is calculated on the basis of the speed acquired in step 1 and the turning angle acquired by the angle acquisition means.

  A positioning apparatus according to a fourth aspect of the present invention is the positioning apparatus according to any one of the first to third aspects, further comprising position reliability determination means for determining the reliability of the position acquired by the position acquisition means, and determining the ambiguity. The means is characterized in that ambiguity is determined according to the determination result of the position reliability determination means.

  The positioning device according to a fifth aspect of the present invention is the positioning apparatus according to the first aspect, wherein the initial value calculating means for calculating the initial value of the ambiguity included in the phase of each carrier wave, and the reliability of the initial value calculated by the initial value calculating means. Initial value reliability determination means for determining, and search range setting means for setting the size of the search range of the ambiguity included in the phase according to the reliability determined by the initial value reliability determination means. The ambiguity determining means is configured to determine ambiguity based on the search range set by the search range setting means.

  The positioning device according to a sixth aspect of the present invention is the positioning device according to the fifth aspect, wherein the search range setting means sets the size of the search range according to the passage of time from the time when the position before positioning is positioned. It is configured.

  A computer program according to a seventh aspect of the present invention is a computer program for causing a computer to function as a means for determining an ambiguity included in a phase of each carrier wave transmitted by a plurality of positioning satellites. A velocity vector calculating means for calculating a velocity vector from its own position and an ambiguity determining means for determining an ambiguity included in the phase using the own position and the calculated velocity vector. Features.

  A positioning method according to an eighth aspect of the present invention is a positioning method by a positioning device that includes a receiving unit that receives each carrier transmitted by a plurality of positioning satellites, and performs positioning based on the phase of each carrier received by the receiving unit. Acquires the position before positioning, calculates the velocity vector from the acquired position, determines the ambiguity included in the phase using the acquired position and the calculated velocity vector, and determines the determined ambiguity It is characterized in that its own position is measured based on it.

  In the 1st invention, 7th invention, and 8th invention, the position (for example, own position measured in the past) of self (positioning object) before positioning is acquired. There are several ways to get your position. For example, a solution (positioning position) determined by RTK-GPS, which will be described later, may be used, or the position of itself can be obtained by communication with a roadside device such as a magnetic nail or an optical beacon. In addition, the one where the precision of a positioning position is higher is desirable. Next, the velocity vector from the acquired own position is calculated. For example, the velocity vector can be calculated by detecting the frequency of the carrier wave transmitted by the positioning satellite and calculating the Doppler shift. Alternatively, the speed vector may be calculated by detecting the speed with a wheel speed sensor and detecting the moving direction with a gyro sensor.

  The ambiguity included in the phase is determined using the acquired position and the calculated velocity vector. More specifically, if the position before positioning is A, the velocity vector is V, and the velocity vector calculation time interval is T, the solution search range center B is obtained by B = A + T · V, The ambiguity (RTK-GPS solution) is determined by limiting the solution search range. The solution search range can be narrowed compared to the case where the solution is obtained by DGPS, and the expected value for obtaining the correct solution is improved. That is, particularly in urban areas, the arrangement of satellites that can receive signals often deteriorates and the influence of multipath is also great. For this reason, the error obtained by the DGPS is likely to have a large error, and an accurate solution is often not obtained because the range for searching for the solution becomes wide. On the other hand, when a velocity vector obtained by Doppler shift is used, Doppler shift is relatively less susceptible to multipath, and noise is very small compared to DGPS using pseudorange. Therefore, the error of the velocity vector is small, and the search range of the solution can be further limited as compared with the case of DGPS. In addition, also when using the speed vector obtained by a wheel speed sensor and a gyro sensor, the search range of a solution can be set according to the precision of the sensor to be used, and ambiguity can be determined. Based on the determined ambiguity, it positions itself. Since the ambiguity is determined after limiting the search range of the solution from past positioning positions and velocity vectors, the ambiguity calculation is faster than the conventional method, and RTK-GPS positioning in urban areas etc. Can greatly increase the time available. And a solution can be calculated by one measurement (1 epoch) of RTK-GPS.

  In the second invention, the velocity vector is calculated based on the Doppler shift of each carrier wave received by the receiving unit. For example, assuming that the frequency of the carrier wave output by the satellite G1 is ν1s and the frequency of the carrier wave measured by itself (positioning target) is ν1r, the Doppler shift (ν1r−ν1s), the known velocity vector of the satellite G1 A certain formula is established between Vg and its own velocity vector V (unknown number). Then, by calculating the satellite position from the ephemeris information received from the satellite and measuring the frequency of the carrier wave from three or more satellites, an equation of an unknown number or more is established, and its own velocity vector V is calculated. Can do. In order to improve time estimation accuracy and satellite position calculation accuracy, it is preferable to receive signals from four or more satellites. Doppler shift is relatively unaffected by multipath and has very little noise compared to DGPS using pseudoranges. For this reason, the error of the velocity vector is small, and the search range of the solution can be limited to a narrow area compared to the case of DGPS.

  In the third aspect of the invention, the speed vector is calculated based on the speed acquired by the speed information acquisition means (for example, wheel speed sensor) and the turning angle acquired by the angle acquisition means (for example, gyro sensor). For example, an accurate speed can be obtained by a wheel speed sensor. If the position of the two points obtained in the past is known, the gyro sensor can accurately determine the turning angle (movement direction) from the direction connecting the two points, and the own velocity vector can be obtained with high accuracy. Can be calculated. Thereby, according to the precision of a wheel speed sensor or a gyro sensor, the search range of a solution can be restricted.

  In the fourth invention, the reliability of the acquired own position is determined. The determination of reliability can be performed as follows, for example. In the case of obtaining the own position by a solution determined by RTK-GPS, in the hatch method, the difference between the double phase difference of the carrier and the value obtained by dividing the pseudorange by the wavelength of the carrier is closer to an integer value. It can be determined that the reliability of the solution (that is, the reliability of the position obtained by the solution) is high. In the case of the LAMBDA method, similarly, the residual between the above-described difference and integer value is used, and the ratio of the smallest residual value to the next smallest residual value among a plurality of solutions is high. It can be determined that the reliability is higher. In any method, the number of satellites that have received a signal, the quality of the signal (ratio of signal to noise), and the like can be added to the determination criteria. Moreover, when calculating | requiring an own position by communication with roadside apparatuses, such as a magnetic nail and an optical beacon, the reliability corresponding to each positioning method can be used. Then, an ambiguity determination method is selected according to the reliability determination result. For example, when the reliability of the position obtained before positioning (that is, in the past) is high, the center of the search range for searching for a solution can be obtained with high accuracy by using the velocity vector from the position. . In this case, the solution search range can be limited to a narrow area, and the possibility that a correct solution can be calculated (particularly, the possibility that a solution can be calculated in one epoch) can be improved. On the other hand, when the reliability of the position obtained before positioning is low, the velocity vector from that position is not used, and the solution obtained by other means, for example, DGPS, is used as the center of the search range. In this case, the solution search range is set wider.

  In the fifth invention, the initial value of the ambiguity included in the phase of each carrier wave is calculated, and the reliability of the calculated initial value is determined. Then, the size of the ambiguity search range is set according to the determined reliability, and the ambiguity is determined based on the set search range. For example, the initial value of the ambiguity is calculated, and the size of the search range for determining the ambiguity is set according to the reliability of the calculated initial value, so the optimal search for searching for the solution A range can be determined.

  In the sixth aspect of the invention, the size of the search range is set in accordance with the passage of time from the time when the own position before positioning is positioned. For example, by increasing the search range with time, even if the reliability of the initial value of the ambiguity (solution) decreases with time, a true solution exists within the search range. It is possible to prevent the situation where the ambiguity cannot be determined.

  In the present invention, the calculation of ambiguity is accelerated, and in particular, the possibility of calculating a solution by one measurement (one epoch) is improved. Thereby, the time which can perform RTK-GPS positioning in an urban area etc. can be increased significantly.

  Hereinafter, the present invention will be described with reference to the drawings illustrating embodiments. FIG. 1 is a schematic diagram showing an outline of an example of a positioning method using a positioning device according to the present invention, and FIG. 2 is a block diagram showing an example of a configuration of a positioning object 100 including a positioning device 10 according to the present invention. . In the positioning method according to the present invention, the position of the positioning target 100 is measured using positioning satellites (GNSS such as GPS, Galileo, and Quasi-Zenith Satellite) G1 to GN, the reference station 300, and the like. The positioning object 100 is, for example, a vehicle or a person. When the positioning object 100 is a vehicle, the positioning object 100 includes the positioning device 10 and the autonomous sensor unit 20 that are positioning devices according to the present invention, and the positioning object 100 is a person. In some cases, a positioning device 10 is provided.

  In the case of GPS, each satellite G1 to GN is equipped with an atomic clock, a 10.23 MHz reference transmitter, etc., and 154 times the reference frequency L1 = 1575.42 MHz and 120 times L2 = 1227.60 MHz. Send the signal. The transmitted signal includes a predetermined positioning code transmitted by the atomic clock at an accurate timing, and how much time (propagation time) is required for the code to reach the receiver. By measuring whether or not, the pseudo-range (propagation time × light speed) with each of the satellites G1 to GN can be measured. In addition, it is not limited to the signal of 2 frequencies, The positioning using only 1 frequency and the positioning using 3 frequencies or more are also possible.

  When finding the solution with RTK-GPS, when using the hatch method, among the satellites G1 to GN, the observable satellites are classified into four main satellites G1 to G4 and the remaining slave satellites (G5 to GN). To do. For example, when there are six observable satellites, the number of slave satellites is two. In addition, when using the LAMBDA method (Least-square Ambiguity Decorrelation Adjustment Method), it is not necessary to distinguish between the primary satellite and the secondary satellite, and there may be four observable satellites G1 to G4. Also, a reference satellite (for example, satellite G1) is determined among the main satellites G1 to G4. By determining the reference main satellite G1, the phase difference of the carrier waves of the other main satellites G2 to G4 is measured based on the phase of the carrier wave of the main satellite G1, and the clock of the positioning device 10 (receiver) The error can be canceled. The position of the positioning object 100 is measured using data of the main satellites G1 to G4. If the hatch method is used, the positioning result is verified using the data of the slave satellites G5 to GN. Note that which satellite is to be the primary satellite or the secondary satellite may be determined by the positioning device 10 based on the satellite orbit information transmitted from the satellite, or the predetermined information is acquired by the positioning device 10 It may be.

  The positioning device 10 includes a processing unit 11 that performs processing such as calculation of a velocity vector, determination of a search range of a solution, determination of a solution (ambiguity), determination of reliability of the solution, and signals from the satellites G1 to GN ( A communication function (for example, a narrow-area communication function, a wireless communication) with a satellite communication unit 12, a reference station 300, a transmission device 200, and a roadside device (not shown) (for example, an optical beacon, a radio beacon, etc.) having a receiving antenna for receiving radio waves) A communication unit 13 having a mid-range communication function such as a LAN, a wide-area communication function, and the like, a storage unit 14 for storing predetermined data, a map database 15, and the like. The satellite communication unit 12 includes a code phase measurement unit 121 that measures pseudo distances from the satellites G1 to GN, a carrier phase measurement unit 122 that measures the phase of the carrier wave of each satellite G1 to GN, and the satellites G1 to GN. And a carrier frequency measuring unit 123 for measuring the frequency of the carrier. The carrier frequency measurement unit 123 includes, for example, an FLL (Frequency Lock Loop) circuit, and can always detect the frequency with high accuracy.

  The autonomous sensor unit 20 includes a wheel speed sensor 201, a gyro sensor 202, and the like. The autonomous sensor unit 20 outputs data detected by the wheel speed sensor 201 and the gyro sensor 202 to the processing unit 11. The processing unit 11 can accurately obtain the speed of the positioning target 100 based on the data detected by the wheel speed sensor 201. Further, if the position of the two points of the positioning target 100 obtained in the past is known, the processing unit 11 accurately obtains the moving direction from the direction connecting the two points using the data detected by the gyro sensor 202. be able to. Accordingly, the processing unit 11 can calculate the velocity vector of the positioning target 100 using the data obtained by the autonomous sensor unit 20. In addition, the process part 11 can calculate the velocity vector of the positioning target 100 by measuring the frequency of the carrier wave of each satellite G1-GN and calculating | requiring a Doppler shift. Details will be described later.

  The reference station 300 has the same function as the satellite communication unit 12, measures the pseudorange with each satellite G <b> 1 to GN, the phase of the carrier wave of each satellite G <b> 1 to GN, and transmits the measurement result to the positioning target 100. By using the measurement result at the reference station 300, the delay time of the radio wave (signal) in the ionosphere or troposphere can be canceled. In addition, it is also possible to perform positioning without installing the reference station 300 by acquiring information on the delay time of radio waves in the ionosphere or troposphere from the main satellites G1 to G4 or from the outside.

  Next, an outline of an example of a positioning method based on a general double phase difference will be described using numerical examples. In addition, the numerical value used here is an example, and is not limited to this. For the carrier wave received by the positioning device 10 and the reference station 300, the difference (double phase difference) between the phase difference of the carrier wave from the main satellite G2 and the phase difference of the carrier wave from the main satellite G1 (reference satellite) is φ21 (t ), And the difference (double phase difference) between the phase difference of the carrier wave from the main satellite G3 and the phase difference of the carrier wave from the main satellite G1 is φ31 (t), and the phase difference of the carrier wave from the main satellite G4 and the main satellite The difference (double phase difference) from the phase difference of the carrier wave from G1 is defined as φ41 (t).

  The carrier wavelength of the main satellites G1 to G4 is λ, and the pseudoranges at the time t between the main satellites G1 to G4 and the reference station 300 are R1r (t), R2r (t), R3r (t), R4r (t ) And the pseudoranges at the time t between the main satellites G1 to G4 and the positioning device 10 are R1u (t), R2u (t), R3u (t), and R4u (t), respectively, Equation (1) Formula (3) is established. Here, N2, N3, and N4 are unknown integers. Since the phase of each carrier wave that can be measured by the positioning device 10 remains in the range of 2π (0 to 360 deg), for example, it cannot be distinguished from the case where the carrier wave phase is 450 deg, 810 deg, and −270 deg. Therefore, the uncertainty is represented by unknown integers N1 to N3 (referred to as ambiguity and integer bias).

  If it is possible to continue capturing radio waves from the main satellites G1 to G4 (if there is no cycle slip), how much the phase of the carrier changes (even if it exceeds 360 deg) while the unknown integers N1 to N3 remain unchanged. You can continue to measure. In other words, once the ambiguity can be calculated (when the solution can be determined), the accurate carrier phase can be obtained continuously. For example, Expressions (4) to (6) can be obtained at the next observation (at the next epoch, for example, time t + Δt). From the expressions (1) to (6), the position of the positioning device 10 (for example, latitude: coordinate X, longitude: coordinate Y, height: coordinate Z) and unknown integers N1 to N3 can be obtained. Equations (1) to (6) can be used for either the Hatch method or the LAMBDA method.

  When the radio wave from the satellite is blocked and a cycle slip occurs, continuous observation is interrupted, and this changes the ambiguity. Therefore, in order to determine the position of the positioning target 100, the ambiguity is calculated again. Must.

  The calculation of the ambiguity is to obtain the unknown integers N1 to N3 as described above, and a large number of equal-phase intersections (a large number of points having the same phase) with the same carrier phase from each satellite, that is, a large number of solutions. Is to find the correct solution (true solution) from the candidates. In other words, solving ambiguity or determining integer bias. One feature of the present invention is that the position of the positioning object 100 before positioning (for example, the past) is acquired, the velocity vector from the acquired position is calculated, and the phase is calculated using the acquired position and the calculated velocity vector. The ambiguity included is determined, and the position of the positioning object 100 is determined based on the determined ambiguity.

  More specifically, assuming that the position of the positioning target 100 before positioning is A, the velocity vector is V, and the time interval for calculating the velocity vector is T, the center B of the solution search range is represented by B = A + T · V. The ambiguity (RTK-GPS solution) is determined by limiting the search range of the solution. The solution search range can be narrowed compared to the case where the solution is obtained by DGPS, and the expected value for obtaining the correct solution is improved. That is, particularly in urban areas, the arrangement of satellites that can receive signals often deteriorates and the influence of multipath is also great. Furthermore, since the noise is relatively large, the solution obtained by DGPS tends to have a large error, and an accurate solution is often not obtained because the range for searching for the solution becomes wide. On the other hand, when using a velocity vector, for example, the Doppler shift is relatively less susceptible to multipath, and noise is very small compared to DGPS using pseudoranges. Therefore, the error of the velocity vector is small, and the search range of the solution can be further limited as compared with the case of DGPS. In addition, also when using the speed vector obtained by the wheel speed sensor 201 and the gyro sensor 202, the search range of a solution can be set according to the precision of the sensor to be used, and ambiguity can be determined. The position of the positioning object 100 is measured based on the determined ambiguity. Since the ambiguity is determined after limiting the search range of the solution from the past positioning position and the velocity vector, in the present invention, the calculation of the ambiguity is accelerated, and in particular, one measurement (one epoch) ) To improve the possibility of calculating the solution. Thereby, the time which can perform RTK-GPS positioning in an urban area etc. can be increased significantly.

  The transmission device 200 can be realized by an optical beacon or a radio beacon having a narrow-area communication function, or a wireless device having a mid-range communication function or a wide-area communication function such as a wireless LAN. The transmission device 200 stores road shape information including altitude information of roads within the positioning area of the positioning object 100, and transmits the stored road shape information to the positioning device 10. The timing of transmission to the positioning device 10 can be, for example, when the positioning target 100 passes through a communication point with the transmission device 200 or enters the communication area.

  When the road shape information is transmitted from the transmission device 200 to the positioning device 10, the positioning device 10 does not need to store the altitude information of the road in advance, and the positioning object 100 moves to an arbitrary point. Even so, the latest information in the vicinity can be acquired. The road shape information can also be stored in the positioning device 10 in advance. In this case, the transmission apparatus 200 is unnecessary, and the system configuration can be simplified.

  In this embodiment, the road includes the ground surface or the ground, and includes a road where a vehicle or a person can move, such as a road on which a vehicle travels, an intersection, a sidewalk where people pass, a vacant lot outside the road, and a parking lot. Shall be.

  FIG. 3 is an explanatory diagram showing an example of an outline of road shape information. In the road shape information, for example, roads are divided at predetermined intervals, nodes are provided in the respective sections, and the horizontal position (X, Y) of the nodes and altitude information Z are associated with each other. As shown in FIG. 3, by setting nodes at predetermined intervals along the road direction, altitude information can be given in a straight line or a curved line corresponding to the horizontal position of the road. The altitude information of the road is, for example, the altitude (height from the geoid surface) of the ground surface (ground) or the ellipsoidal height (height from the earth spheroid). The geoid surface is a surface virtually defined as an isogravity surface by extending the sea level where unevenness occurs according to the magnitude of gravity to the land.

  FIG. 4 is an explanatory diagram showing another example of the outline of road shape information. In this case, the road shape information is obtained by dividing a road into a matrix by rectangular small areas, providing a node in each small area, and associating the horizontal position (X, Y) of the node with the altitude information Z. is there. As shown in FIG. 4, altitude information can be given in a plane corresponding to the horizontal position of the road.

  The altitude information of the road can be measured in advance at each point (node) in the positioning area using RTK-GPS in a time zone where the number of observable satellites and the satellite arrangement are good. In addition, when measuring altitude information such as roads using RTK-GPS, a value obtained by subtracting the height of RTK-GPS from the road surface from the altitude measured by RTK-GPS is used as altitude information on the ground surface. Thereby, the altitude information of the road can be obtained with high accuracy. The interval between the nodes can be set as appropriate, but can be measured, for example, at an interval of 5 m. Further, altitude data included in the map data can be used as the altitude information of the road, and in this case, it can be acquired relatively easily.

  FIG. 5 is an explanatory diagram showing an example of the data structure of road shape information. As shown in FIG. 5, the road shape information includes fields such as a node ID, a horizontal position (latitude and longitude) of the node, altitude information, and a method for measuring altitude information. For example, the horizontal position of the point whose node ID is a1 is (X1, Y1), the altitude information at that point is Z1, and the altitude information is measured by RTK-GPS. In addition, the data structure of road shape information is an example, Comprising: It is not limited to this.

  When the transmission device 200 transmits road shape information to the positioning device 10, it is possible to determine how much road shape information in the region is to be transmitted according to the purpose or application. For example, when the transmission apparatus 200 is configured with an optical beacon and a vehicle that moves from the upstream of the intersection toward the intersection is the positioning target 100, when the position of the vehicle to the intersection is accurately measured, communication with the optical beacon is performed. Road shape information in a region from the position to the vicinity of the intersection can be transmitted.

  When the positioning device 10 performs positioning in a relatively wide area, road shape information in the area may be transmitted. In addition, when it is possible to repeatedly perform communication between the positioning device 10 and the transmission device 200, when the positioning device 10 estimates its approximate horizontal position as described later, the estimated horizontal position data is stored. The transmission apparatus 200 can also transmit road shape information in the vicinity of the received horizontal position (for example, within several meters) to the positioning apparatus 10 every time it is received. Thereby, the amount of communication can be reduced.

  Next, the operation of the positioning device 10 will be described. FIG. 6 is a flowchart showing the procedure of the positioning process of the positioning device 10. The processing unit 11 or the like may be configured by dedicated hardware or a semiconductor module (chip), or may be configured by a CPU or a RAM, and the CPU may execute a program code indicating a predetermined processing procedure to execute the processing unit 11. It is also possible to realize the function. Thereby, the positioning device according to the present invention can be realized by the program code. The program code can also be recorded on a recording medium such as an optical disk, a magnetic disk, or a semiconductor memory, and the program code recorded on such a recording medium is loaded into an information processing apparatus including a CPU and a RAM. The positioning device according to the present invention can also be realized. Note that the process shown in FIG. 6 is based on the use of the Hatch method. In the LAMBDA method, the solution search range is substantially controlled by controlling the probability that the solution is found at a position far from the center of the solution search range by parameter setting in step S15 described later. Other steps are the same as in the Hatch method. In the example of FIG. 6, the processes in steps S11 to S13 are processes in the previous stage for determining ambiguity, and the processes after step S14 are the main processes for determining ambiguity. .

  The processing unit 11 measures the position of the positioning object 100 as a previous step for determining ambiguity (S11). Here, the positioning of the position may be a method of positioning its own position by communication with a roadside device such as a magnetic nail or an optical beacon, or a method of obtaining the position by calculating a solution using RTK-GPS. Hereinafter, the positioning of the position as the previous stage by calculating the solution using RTK-GPS will be described.

  Step S11 described above can be performed in the following two steps using, for example, RTK-GPS. In the first step, the estimated position of the positioning object 100 is obtained by using the measured pseudo distance by the same method as a general DGPS receiver. Since the error of the estimated position obtained here is relatively large (when the value corresponding to the standard deviation of the estimated position is σp, σp is about 60 cm, for example), the more accurate position in the next second step. Ask for. In the second step, the search range is limited around the position obtained in the first step, and a solution is obtained by RTK-GPS using the phase of the carrier wave.

  FIG. 7 is an explanatory diagram showing an example of positioning by DGPS. The accuracy of the receiver clock mounted on the positioning device 10 is generally lower than the atomic clock mounted on each satellite G1 to G4, and is not synchronized with the clock mounted on each satellite. Therefore, the distances r1 to r4 with the satellites G1 to G4 measured by the positioning device 10 are not true distances, but are pseudo distances including errors due to time errors in the true distances. Since the positions of the satellites G1 to G4 are known in advance, the clock error of the receiver is corrected using the fourth satellite, the true distance between the three satellites and the receiver is obtained, and the three satellites are centered. The three spheres having a true radius can be drawn, and the intersection of the three spheres can be obtained as the estimated position (for example, latitude and longitude) of the positioning object 100.

  When using the distance to each of the satellites G1 to G4, the measurement accuracy depends on the performance of the receiver and the surrounding environment, and is generally about 1 m to 10 m. The estimated position is, for example, a circle with a diameter of 1 m to 10 m. It can be obtained as an area. Note that the shape of the region is not limited to a circular shape, and may be a rectangular shape. Also, in this method, particularly in urban areas, it is strongly influenced by the reflection of signals from buildings such as buildings, so even if an advanced multipath removal algorithm is used, it corresponds to the standard deviation of the estimated position error. The value σp is about 60 cm. However, it can be used to obtain the approximate position of the positioning object 100.

  As a second step, the search range is limited around the position obtained by DGPS, and a solution is obtained by RTK-GPS using the phase of the carrier wave. In RTK-GPS, a solution cannot be uniquely obtained from data obtained by one measurement (one epoch). That is, for example, if the number of satellites that can receive a carrier wave is four, the unknown number is a total of six of three positions (x, y, z) of the positioning target 100 and three integer ambiguities. This is because there are three equations, but there are three equations obtained in one measurement. Therefore, an area in which a solution exists is limited in advance, and the most likely solution is obtained in the limited search range.

  As a method of restricting the area where the solution exists, for example, (1) restriction around the position obtained by DGPS in the first step described above, and (2) solution using road shape information including road altitude information is used. Limit the search range. Note that a plurality of these limiting methods can be combined. The method using the speed vector will be described later, and the case where the road shape information is used will be described.

  FIG. 8 is a conceptual diagram showing an example in which the solution search range is limited using road shape information. It is assumed that the positioning device 10 estimates the approximate position (latitude, longitude) by the method of the example of FIG. The positioning device 10 acquires road altitude information from road shape information corresponding to an estimated position (which can be specified as a certain area as shown in FIG. 8). In FIG. 8, as an example, it is assumed that altitude information Z11, Z12, Z13, and Z14 at four points can be acquired. If there is no altitude information corresponding to the estimated position, the altitude information may be calculated by performing interpolation processing using the altitude information closest to the horizontal position.

  The positioning device 10 assumes that the errors of the altitude information Z11 to Z14 are e11 to e14 with respect to the acquired altitude information Z11 to Z14, respectively. An area between the minimum height and the maximum height among −e14 and Z14 + e14 is specified as a solution (ambiguity) search range S. The example of FIG. 8 schematically shows the solution search range S obtained in this way. Thereby, the positioning apparatus 10 can limit to the solution which exists in the search range S among many solution candidates, and the solution candidate which exists outside the search range S is not a correct solution (ambiguity). Can be excluded. Thereby, ambiguity can be solved at high speed.

  When the positioning device 10 is mounted on a vehicle or carried by a person so that it can be carried, the height position of the satellite communication unit 12 (more precisely, the receiving antenna) depends on the height of the receiving antenna. It will be added (height from the road surface). For this reason, in order to perform positioning with higher accuracy, it is necessary to specify the height position of the receiving antenna. Further, the height position of the receiving antenna differs depending on where the vehicle is attached or how a person holds it, and it is preferable to consider an error in the height position of the receiving antenna. Therefore, the height range of the receiving antenna can be determined with higher accuracy by setting the accuracy information (reliability) of the altitude information in association with the altitude information of the road.

  As a method of limiting the solution search range by the above-described limiting method and finally determining a solution (ambiguity) from finite solution candidates, for example, there is a Hatch method, and the method of the present invention is applied. Can do. Also, when using the LAMDA method, altitude information can be used for testing the solution.

  Next, the processing unit 11 determines the reliability of the measured position (S12). The determination of the reliability of the measured position can be performed as follows, for example. In the case of the hatch method, when the measured position is obtained by a solution determined by RTK-GPS, the difference between the double phase difference of the carrier and the value obtained by dividing the pseudorange by the wavelength of the carrier is close to an integer value. It can be determined that the reliability of the solution (that is, the reliability of the position obtained by the solution) is high. In the case of the LAMBDA method, similarly, the residual between the above-described difference and integer value is used, and the ratio of the smallest residual value to the next smallest residual value among a plurality of solutions is high. It can be determined that the reliability is higher. In addition, when the position is measured by communication with a roadside device such as a magnetic nail or an optical beacon, reliability corresponding to each roadside device can be used.

  The processing unit 11 determines whether or not the reliability is high (S13). If the reliability is not high (NO in S13), the processing from step S11 is continued. In this process for determining ambiguity (process after step S14), since the solution is determined by one measurement (one epoch), a highly reliable solution (position) is used as the position before positioning. This is because it is important to use.

  When the reliability is high (YES in S13), the processing unit 11 calculates a velocity vector (S14). That is, only when it is determined that the reliability is high, the obtained position and velocity vector are used in the next and subsequent ambiguity determination. For example, when the reliability of the position obtained before positioning (that is, in the past) is high, the center of the search range for searching for a solution can be obtained with high accuracy by using the velocity vector from the position. . In this case, the solution search range can be limited to a narrow area, and the possibility that a correct solution can be calculated (particularly, the possibility that a solution can be calculated in one epoch) can be improved.

  The velocity vector can be calculated, for example, by detecting the frequency of the carrier wave transmitted by each of the satellites G1 to GN, thereby calculating the Doppler shift and calculating its own velocity vector. Alternatively, the speed vector may be calculated by detecting the speed with a wheel speed sensor and detecting the moving direction with a gyro sensor. First, an example in which a velocity vector is calculated using Doppler shift will be described.

  FIG. 9 is an explanatory diagram showing an example of calculating a velocity vector using Doppler shift. As shown in FIG. 9, it is assumed that the frequency of the carrier wave output from the satellite G1 is ν1s, and the positioning object 100 receives the carrier wave of the satellite G1 and measures the frequency of the carrier wave ν1r. Further, the known velocity vector of the satellite G1 is Vg1 (vx1, vy1, vz1), and the unknown velocity vector of the positioning object 100 is V (vx, vy, vz).

  When the relative velocity vector of the positioning target 100 with respect to the satellite G1 is Vr1, the velocity vector Vr1 can be expressed by Expression (7).

  When the angle between the direction from the positioning target 100 toward the satellite G1 and the direction of the velocity vector Vr1 is θ1, Equation (8) is established and Equation (9) is obtained. Here, the magnitude of the relative velocity vector Vr1 is approximated as being sufficiently smaller than the light velocity c.

  Similarly, the frequency of the carrier wave output by the satellite G2 is ν2s, and the frequency of the carrier wave measured by the positioning object 100 by receiving the carrier wave of the satellite G2 is ν2r. The known velocity vector of the satellite G2 is Vg2 (vx2, vy2, vz2), the relative velocity vector of the positioning target 100 with respect to the satellite G2 is Vr2, and the direction from the positioning target 100 toward the satellite G2 and the velocity vector Vr2 When the angle formed with the direction is θ2, Equation (10) is obtained.

  Similarly, the frequency of the carrier wave output from the satellite G3 is denoted by ν3s, and the frequency of the carrier wave measured by the positioning object 100 receiving the carrier wave of the satellite G3 is denoted by ν3r. The known velocity vector of the satellite G3 is Vg3 (vx3, vy3, vz3), the relative velocity vector of the positioning target 100 with respect to the satellite G3 is Vr3, and the direction from the positioning target 100 toward the satellite G3 and the velocity vector Vr3 When the angle formed with the direction is θ3, the equation (11) is obtained.

  If the unit vector in the direction from the positioning object 100 toward the satellite G1 is E (ex1, ey1, ez1), the left side of the equation (9) is the inner product of the velocity vector Vr1 and the unit vector E. Is established. Similarly, equations (10), (11) to (12) can be finally obtained as many as the number of satellites, and the satellite position is calculated from the ephemeris information received from the satellite. If the Doppler shift of the carrier wave from the satellite can be measured (the frequency of the carrier wave can be measured), the velocity vector V of the positioning object 100 can be obtained. In order to increase the accuracy of time estimation and increase the accuracy of satellite position calculation, it is preferable to receive signals from four or more satellites. Doppler shift is relatively unaffected by multipath and has very little noise compared to DGPS using pseudoranges. For this reason, the error of the velocity vector is small, and the search range of the solution can be limited as compared with the case of DGPS.

  When calculating the speed vector of the positioning object 100 using the wheel speed sensor and the gyro sensor, for example, the speed vector is calculated based on the speed detected by the wheel speed sensor and the moving direction detected by the gyro sensor. An accurate speed can be obtained by the wheel speed sensor. Then, if the positions of the two points of the positioning target 100 obtained in the past are known, the moving direction from the direction connecting the two points can be accurately obtained by the gyro sensor, and the velocity vector of the positioning target 100 can be obtained with high accuracy. Can be calculated. Thereby, the search range of the solution can be limited.

  The processing unit 11 determines a solution search range using the position of the positioning object 100 measured in steps S11 to S13 and the calculated velocity vector (S15).

  FIG. 10 is an explanatory diagram showing an example of a method for determining a solution search range. In FIG. 10, a point A indicates the position of the positioning object 100 before positioning (past), and is a highly reliable positioning position obtained in steps S11 to S13, for example. If the calculated velocity vector is V and the time interval for calculating the velocity vector is T, the center B of the solution search range is obtained by B = A + T · V. The distance rd from the center B of the search range to the boundary of the search range is, for example, standard deviation σd × 3 (significant level 99%), standard deviation σd × 2 (significant level 95%), etc. A corresponding value can be used, and in this case, for example, rd = 3σd = 45 cm. The distance rd can be stored in the storage unit 14 in advance. The search range may have various shapes such as a spherical shape, a circular shape, a rectangular shape, a cubic shape, and a rectangular parallelepiped shape.

  The solution search range can be narrowed compared to the case where the solution is obtained by DGPS, and the expected value for obtaining the correct solution is improved. For example, in the case of DGPS, when a value corresponding to the standard deviation of error σp × 3 (significance level 99%) is used, σp = about 60 cm, so the solution search range is about 180 cm. On the other hand, when the velocity vector obtained by the above-mentioned Doppler shift is used, it is about 45 cm. At the same significance level, the radius of the search range is reduced by about 135 cm, and the search range is smaller than that in the case of DGPS. It will be about 1/64.

  That is, the solution obtained by DGPS is likely to have a large error due to the influence of multipath, and an accurate solution cannot be obtained in many cases because the range for searching for the solution becomes wide. On the other hand, when a velocity vector obtained by Doppler shift is used, since the Doppler shift is not easily affected by multipath, the error of the velocity vector is small and the search range of the solution can be further limited as compared with the case of DGPS. In addition, also when using the speed vector obtained by a wheel speed sensor and a gyro sensor, the search range of a solution can be set according to the precision of the sensor to be used, and ambiguity can be determined. Based on the determined ambiguity, it positions itself. Since the ambiguity is determined after limiting the search range of the solution from the past positioning position and velocity vector, the ambiguity calculation is faster than the conventional method, especially one measurement (one epoch). Improve the possibility of calculating the solution with. Thereby, the time which can perform RTK-GPS positioning in an urban area etc. can be increased significantly.

  FIG. 11 is an explanatory diagram schematically showing how the solution search range is limited. As described above, there are many points (solution candidates) where the phases of the carrier waves from the main satellites G1 to G4 are equal. In the example of FIG. 11, the equiphase surfaces of the carrier waves from the main satellites G1 and G2 are schematically shown. A point where isophase planes intersect (indicated by a circle in the figure) is a solution candidate. By using the velocity vector, it is possible to limit the number of solution candidates to those existing in the region indicated by the solution search range S. The same applies to the other main satellites G3 and G4.

  The processing unit 11 calculates a solution from the solution candidates within the determined search range (S16), and determines the reliability of the solution (S17). The solution reliability can be determined using the same method as in step S13. That is, when calculating the solution with RTK-GPS, when using the Hatch method, the closer the difference between the double phase difference of the carrier and the value obtained by dividing the pseudorange by the wavelength of the carrier is closer to the integer value, the more reliable the solution is. It can be determined that the property (that is, the reliability of the position obtained from the solution) is high. Similarly, when the LAMBDA method is used, similarly, the residual between the above-described difference and the integer value is used, and the ratio between the smallest residual value of the plurality of solutions and the next smallest residual value is obtained. It can be determined that the higher the reliability, the higher the reliability.

  When the reliability of the solution is not high (NO in S18), the processing unit 11 continues the processing from step S11. If the reliability of the solution is not high, the solution cannot be obtained using the velocity vector by the Doppler shift at the next measurement (epoch), so the process returns to the previous stage.

  When the reliability of the solution is high (YES in S18), the processing unit 11 measures the position (S19) and determines whether or not to continue the process (S20). When the process is continued (YES in S20), the processing unit 11 continues the process after step S14 to calculate the solution using the velocity vector by the Doppler shift even at the next measurement (epoch), and does not continue the process. (NO in S20), the process is terminated.

  In the example of FIG. 6 described above, a schematic flow for calculating a solution has been described. Hereinafter, the processing when the Hatch method is used will be described in more detail.

  12 to 15 are flowcharts showing the procedure of the positioning process by the Hatch method of the positioning device 10. In the flowchart of FIG. 6, locations corresponding to steps S11 to S13 are steps S101 to S113 of FIGS. 12 and 13. Further, in the flowchart of FIG. 6, a portion corresponding to step S <b> 14 and subsequent steps corresponds to step S <b> 114 and subsequent steps in FIGS. 14 and 15.

  The processing unit 11 measures the carrier phase and pseudorange of the signals from the main satellites G1 to G4 (S101), and receives the carrier phase and pseudorange of the signals from the main satellites G1 to G4 measured by the reference station 300 ( S102). The processing unit 11 calculates each double phase difference of the carrier phase (S103).

  If the carrier phases from the main satellites G1 to G4 measured by the positioning device 10 are Φ1u to Φ4u, respectively, and the carrier phases from the main satellites G1 to G4 measured by the reference station 300 are Φ1r to Φ4r, respectively, the main satellite G1 (reference satellite) The carrier phase difference (double phase difference) of each of the main satellites G2 to G4 with reference to the carrier phase of) can be expressed by Equations (13) to (15). Here, ▽ Δ is an operator (operator) for calculating the double phase difference. By using the double phase difference, it is possible to cancel the radio wave delay time in the ionosphere or troposphere and the clock error of the receiver of the positioning device 10.

  By using the operator ▽ Δ, the expressions (13) to (15) can be expressed as the expression (16). Here, i = 2, 3, and 4. Hereinafter, an operator for calculating the double phase difference is defined as in Expression (16).

  In the present embodiment, the calculation of the double phase difference is performed with respect to the phase difference (referred to as a wide lane) between the L1 signal (1575.42 MHz) and the L2 signal (1222.70 MHz) of each of the main satellites G1 to G4. That is, the wide lane Φ (wide) is defined by the phase difference between L1 and L2 as in Expression (17), and Expression (18) is established.

  Equation (19) is established for the double phase difference of the wide lane. Here, i = 2, 3, and 4.

  In Expression (19), Ri is a pseudorange obtained using the code phase between the receiver of the positioning device 10 or the reference station 300 and each of the main satellites G1 to G4, and can be expressed by Expression (20). N (wide) is ambiguity and an integer value, and can be expressed by equation (21). Also, λ (wide) is the wavelength of the wide lane and can be expressed by equation (22). Since the L1 signal is 1575.42 MHz and the L2 signal is 1222.70 MHz, λ (wide) is 0.863 m. Become.

  Since there are many points where the double phase difference of the wide lane becomes equal for each phase 2π, the expression (19) is expressed as an expression including ambiguity (solution candidate). . The positioning apparatus 10 calculates the phase difference of each carrier wave received by the receiver, and acquires the phase difference of each carrier wave measured by the reference station 300 from the reference station 300. The positioning device 10 determines the ambiguity based on the calculated phase difference and the acquired phase difference, and uses the known position information of the reference station 300, the position information of the satellite, and the distance to the satellite, Can be determined. Thus, it can be seen that when the ambiguity is obtained, the position of the receiver is obtained.

  The processing unit 11 determines whether or not a solution candidate remains (S104). If no solution candidate remains (NO in S104), the processing unit 11 calculates an initial value of ambiguity from the measured pseudorange. Then, the reliability of the calculated initial value is calculated (S106). Whether or not a solution candidate remains is determined by determining the true solution when a plurality of solution candidates are obtained in the most recent process but cannot be determined as one solution (true solution) This is for repeating the process. Accordingly, no solution candidates remain in the process of obtaining the solution candidates first.

  The initial value of ambiguity can be calculated by, for example, Expression (23).

  In Expression (23), Φ (wide) and Ri can be measured, and λ (wide) can be calculated by Expression (22). Equation (23) is the difference between the fraction of the carrier wave (including the integer part and the fractional part) present in the value obtained by taking the double phase difference with ▽ ΔRi and the measured carrier wave phase difference (within 0 to 2π). Then, the nearest integer value is obtained as the initial value of the ambiguity. As a result, a value close to the solution (ambiguity) to be calculated can be obtained as an initial value, and the determination rate of the solution can be improved.

  In the above-described calculation of the initial value of ambiguity, the pseudo distance is used, but other measurement values can also be used. For example, when the positioning target 100 is a vehicle, when it is possible to estimate that the position of the receiver antenna is within a range of about 1 m that can be communicated with the optical beacon by communication with the optical beacon, the estimated value and each main satellite The absolute distance from G1 to G4 can be used instead of the pseudo distance. This makes it possible to calculate a highly accurate initial value.

  Further, the reliability of the initial value of the ambiguity can be calculated according to the measurement accuracy of the measurement value used when calculating the initial value of the ambiguity.

  FIG. 16 is an explanatory diagram showing an example of the reliability of the initial value of the ambiguity. As illustrated in FIG. 16, an evaluation value table in which a positioning method and reliability evaluation values are associated with each other can be stored in the storage unit 14. For example, at the time of communication with an optical beacon, the estimated position of the positioning device 10 can be limited to the communication area with the optical beacon, and the position can be measured with high accuracy, so the evaluation value is set to 90.

  In addition, when the solution is obtained by RTK-GPS (carrier synchronization type), the evaluation value is set to 90 because the influence of multipath is small and the position can be measured with high accuracy. Further, when a solution is obtained by GPS (code synchronization type), for example, an evaluation value is set in a range of 50 to 70 according to a reception situation. For example, when the number of received satellites is 4, the DOP (dilution of precision) value is large, the signal strength and the SN ratio are small, the evaluation value is set to 50, the number of received satellites is 8, the signal strength, When the SN ratio is large, the evaluation value is set to 70. Considering the degree of distortion of the correlation function of the received signal as the reception status, a small evaluation value can be set when the distortion is large, and a large evaluation value can be set when the distortion is small.

  Also, when communicating with a magnetic nail buried on the road surface or a radio with a predetermined transmission point, the evaluation value is set to 85, and when communicating with a DSRC (for example, an ETC toll gate), Although there is a spread, the estimated position is not greatly mistaken within that range, so the evaluation value is set to 75. Note that the above evaluation value is merely an example, and the present invention is not limited to this.

  The processing unit 11 determines an ambiguity search range based on the calculated reliability (S107). When the solution candidates remain (YES in S104), the processing unit 11 performs the process of step S107. When two or more solution candidates remain, the initial value of ambiguity is not calculated again by skipping steps S105 and S106. This is because if there are two or more candidate solutions, there is a high possibility that a true solution exists in the vicinity of the initial value calculated earlier. This is because a true solution can be obtained efficiently by obtaining.

  The processing unit 11 calculates a solution for each estimated candidate using the data of the main satellite (S108). The estimation of the ambiguity candidate can be performed by Expression (24). That is, the true ambiguity (true solution) can be assumed to be within the range represented by the equation (24).

  Here, σp is a value corresponding to the reliability of the initial value of the ambiguity, and can be decreased as the reliability is higher, and can be increased as the reliability is lower. That is, if the initial value of ambiguity can be estimated with high accuracy, σp can be set to a smaller value. As σp, a value corresponding to the standard deviation of the initial value (estimated value) of ambiguity can be used, and can be set to 60 cm, for example. Further, k can be a value such as k = 2, 3, etc., and corresponds to a significance level of 95% and 99%, respectively. Note that the value of σp can be stored in the storage unit 14 in advance.

  As an example, when k = 3 and σp = 60 cm, the wavelength λ (wide) of the wide lane is about 86 cm, so that it can be considered that a true solution exists between the initial value of ambiguity ± 2 cycles. it can. That is, for each of the main satellites G2 to G4 excluding the reference main satellite G1, there are five solution candidates, and the total number of solution candidates is 5 to the third power = 125.

  The positioning device 10 performs a test to determine a true solution from the solution candidates. The processing unit 11 verifies each solution (solution candidate) using the data of the slave satellites G5 to GN (S109).

  The verification process using the data of the slave satellites G5 to GN can be performed as follows. That is, the ambiguity is determined based on the carrier wave transmitted by each of the main satellites G1 to G4, and the position of the positioning device 10 (self) is determined using the determined ambiguity. The positioning device 10 calculates the distance between the satellites G5 to GN and the receiving unit using the position of the positioning device 10 and measures the phase of the carrier wave transmitted by the satellites G5 to GN. The positioning device 10 calculates the value represented by the equation (25) based on the calculated distance and the measured phase, and tests the reliability of the previously determined ambiguity. Here, i = 5 to N. This is because the true solution among the solutions obtained from the data of the main satellites G1 to G4 is considered to be a value close to the integer value of the value calculated by the equation (25) using the data of the slave satellites G5 to GN. It is.

  The solution (ambiguity) obtained for each candidate, that is, the difference between the integer value and the integer value closest to the value calculated by Equation (25) is a predetermined threshold (for example, 0.1, 0.2, etc.) ) Is discarded, the solution (ambiguity) is discarded, and if it is smaller than a predetermined threshold, it is left as a true solution candidate.

  As described above, by performing the verification process, the number of candidates can be narrowed down from a large number of solution candidates, and determination of the solution (ambiguity) can be performed at high speed and with high accuracy. .

  The processing unit 11 determines whether there is a solution candidate (S110). If there is a solution candidate (YES in S110), the processing unit 11 determines whether there is one solution candidate (S111). When there is one solution candidate (YES in S111), the processing unit 11 determines the remaining one candidate as a true solution (S112), and determines the position of the positioning target 100 based on the determined solution. (S113).

  On the other hand, when there is no solution candidate (NO in S110), or when there is not one solution candidate (NO in S111), the processing unit 11 continues the processing after step S101 in the next epoch. Through the above processing, the processing unit 11 can acquire a positioning position with high reliability as the previous stage.

  The processing unit 11 measures the carrier phase and frequency of signals from the main satellites G1 to G4 (S114), and calculates the velocity vector using the Doppler shift (S115). The processing unit 11 receives the carrier phase of the signals from the main satellites G1 to G4 measured by the reference station 300 (S116), and calculates each double phase difference of the carrier phase (S117). In this case, equations (13) to (22) are established.

  The processing unit 11 determines whether or not a solution candidate remains (S118). If no solution candidate remains (NO in S118), the previous positioning position (the positioning position obtained in step S113). And the initial value of ambiguity is calculated using the calculated velocity vector (S119). Here, the initial value of the ambiguity is, for example, the point B in the example of FIG. 10, and can be the center of the solution search range.

  The processing unit 11 calculates the reliability of the calculated initial value (S120). Note that the reliability calculated in step S120 is higher than the reliability calculated in step S106. When a velocity vector obtained by Doppler shift is used, Doppler shift is relatively less susceptible to multipath, and noise is very small compared to DGPS using pseudorange. This is because there is little error in the velocity vector.

  The processing unit 11 determines an ambiguity search range based on the calculated reliability (S121). Note that the search range determined in step S121 is narrower than the search range determined in step S107. When a velocity vector obtained by Doppler shift is used, Doppler shift is relatively less susceptible to multipath, and noise is very small compared to DGPS using pseudorange. This is because there is little error in the velocity vector.

  When the solution candidate remains (YES in S118), the processing unit 11 performs the process of step S121. The processing unit 11 calculates a solution for each estimated candidate using the data of the main satellite (S122). The ambiguity candidate can be estimated by Expression (26). That is, the true ambiguity (true solution) can be assumed to be within the range expressed by the equation (26). In addition, the range represented by Formula (26) is narrower than the range represented by Formula (24).

  Here, σd is a value corresponding to the reliability of the initial value of the ambiguity, and can be decreased as the reliability is higher and can be increased as the reliability is lower. That is, when the initial value of ambiguity can be estimated with high accuracy, σd can be set to a smaller value. As σd, a value corresponding to the standard deviation of the initial value (estimated value) of ambiguity can be used, and can be set to 15 cm, for example. Further, k can be a value such as k = 2, 3, etc., and corresponds to a significance level of 95% and 99%, respectively.

  The positioning device 10 performs a test to determine a true solution from the solution candidates. The processing unit 11 verifies each solution (solution candidate) using the data of the slave satellites G5 to GN (S123).

  The verification process using the data of the slave satellites G5 to GN can be performed in the same manner as S109 using Expression (25). This can be done as follows. As described above, by performing the verification process, the number of candidates can be narrowed down from among a large number of solution candidates, and determination of the solution (ambiguity) is performed at high speed (in one epoch) and with high accuracy. It becomes possible.

  The processing unit 11 determines whether there is a solution candidate (S124). If there is a solution candidate (YES in S124), the processing unit 11 determines whether there is one solution candidate (S125). When the number of solution candidates is one (YES in S125), the processing unit 11 determines the remaining one candidate as a true solution (S126), and determines the position of the positioning target 100 based on the determined solution. (S127).

  The processing unit 11 determines whether or not to continue the processing (S128). When the processing is continued (YES in S128), the processing unit 11 determines whether or not a predetermined time has elapsed since the determination of the solution (S129).

  In the example of FIG. 10, when the point B is obtained from the highly reliable positioning position A by using the velocity vector V, the method is not limited to the method of calculating the velocity vector once, for example, at the time t0. When A is measured and velocity vectors V (t1), V (t2), and V (t3) are calculated every time after time t1, t2, and t3, point B is expressed as B = A + (t1-t0). V (t1) + (t2-t1) .V (t2) + (t3-t2) .V (t3). In such a case, the reliability of the sum of velocity vectors decreases with time. Therefore, a solution limiting method can be selected in accordance with the elapsed time after a highly reliable solution is obtained. That is, when a predetermined time has elapsed since the determination of the solution, the solution is limited by DGPS without using a velocity vector due to Doppler shift, and when the predetermined time has not elapsed since the determination of the solution, the search range of the solution Therefore, it is possible to determine the search range of the solution using the velocity vector due to the Doppler shift.

  In addition, with the passage of time from the time when the position before positioning is measured, the search range is increased (for example, when the elapsed time is tripled, σd is tripled as the route), thereby elapse of time. Even if the reliability of the initial value of the ambiguity (solution) decreases according to the situation, the true solution can be made to exist within the search range and the ambiguity cannot be determined. Can be prevented.

  If the predetermined time has not elapsed since the determination of the solution (NO in S129), the processing unit 11 continues the process from step S114, and if the predetermined time has elapsed since the determination of the solution (YES in S129), the processing unit 11 The processing after step S101 is continued. If there is no solution candidate (NO in S124), or if there is not one solution candidate (NO in S125), the processing unit 11 continues the processing from step S101 onward in the next epoch. When the process is not continued (NO in S128), the processing unit 11 ends the process.

  In the above example, the solution is verified using the data of the slave satellites in steps S109 and S123. However, the present invention is not limited to this, and each solution is verified using the estimated position. Alternatively, each solution can be tested using altitude information.

  When each solution is tested using the estimated position, the estimated position is, for example, a position vector (latitude, longitude, or baseline vector from the reference station 300) using a pseudorange obtained by a GPS (code synchronous) receiver. Then, the position vectors calculated based on the solution (ambiguity) obtained for each candidate are compared, and those that are far apart are discarded. For example, it can be tested using equation (27).

  Here, X (r) is a position vector obtained using a pseudorange, and X (w) is a position vector obtained using a wide lane in the above positioning processing procedure. The symbol || of the absolute value indicates the norm in the horizontal direction, and σ2 is a value corresponding to the standard deviation in the horizontal direction of the difference between the position obtained by the pseudo distance and the position obtained by using the wide lane. Further, k = 2 and 3 correspond to significance levels of 95% and 99%, respectively.

  For example, σ2 can be expressed by Expression (28), and can be determined in consideration of the satellite arrangement at the time of measurement. In Equation (28), RHDOP is RDOP (relative dilution of precision) in the horizontal direction at the time of measurement. That is, when the satellite arrangement is not good (the dispersion of the satellite arrangement is small), the value of RHDOP is large and the accuracy of X (r) is low, so the difference between the measurement results of X (r) and X (w) Increase the tolerance. The processing unit 11 discards the solution when the expression (27) is satisfied, and leaves it as a true solution candidate when the expression (27) is not satisfied.

  Also, when testing each solution using altitude information, specifically, assuming that the altitude obtained using the wide lane is Z (w), if the equation (29) or (30) is satisfied, the solution is If it is discarded and Equation (29) and Equation (30) are not satisfied, it is left as a true solution candidate. In Expressions (29) and (30), Zmax and Zmin are, for example, an upper limit value and a lower limit value that specify the height position range.

  Next, processing when the LAMBDA method is used as a calculation method for calculating a solution will be described. As preconditions, there are four visible satellites G1 to G4, three observation equations obtained as a result of observation with the four satellites G1 to G4, the unknown number is the position to be measured (x, y, z) and the integer ambi It is assumed that there are a total of six Guits N1, N2, and N3.

  First, the initial position (x0, y0) of the positioning object 100 is determined by any method such as a solution determined by DGPS (a method using a pseudorange) or a solution obtained from a previous positioning position and velocity vector by RTK-GPS. , Z0).

  Next, the integer ambiguities N1, N2, and N3 are integers, but if the initial position (x0, y0, z0) is fixed by removing the constraint and accepting the real solution, the three observation equations N1r, N2r, and N3r as real numbers can be obtained. These are called float solutions.

  Next, true integer solutions N1, N2, and N3 (fixed solutions) are searched from around N1r, N2r, and N3r. For this purpose, it is assumed that a weighting matrix Wn is defined and given as an integer least square problem and the cost function C (N) represented by the equation (31) is minimized. Here, N can be represented by Expression (32), and Nr can be represented by Expression (33).

  Further, each element of the matrix Wn is set to a smaller value as the standard deviation of the measured value is larger. In Nr and Wn thus determined, N (an integer ambiguity) that minimizes the cost function C (N) is determined. For a specific method for obtaining the ambiguity N of an integer value, for example, refer to a non-patent document translated from the Japan Voyage Society GPS Study Group, “Sophisticated GPS Basic Concept / Positioning Principle / Signal and Receiver”. it can.

  In this case, as the element of the matrix Wn is smaller, that is, as the standard deviation σn is larger, the cost function c (N) may be minimized at a value (N1, N2, N3) far from (N1r, N2r, N3r). Becomes higher. That is, the solution search range can be substantially adjusted by setting each element of the matrix Wn to a smaller value as the standard deviation of the measured value is larger. For example, “The LAMBDA method for integer ambiguity estimation: implementation aspects”, Paul de Jonge and Christian Tiberius, August 1996, can be referred to.

  In the LAMBDA method, it is not necessary to distinguish between the primary satellite and the secondary satellite, and the formula (24), formula (25), formula (27), formula (28), etc. used in the hatch method are not necessary. is there.

  Next, experimental results performed by applying the present invention will be described. FIG. 17 is an explanatory diagram showing an example of a positioning result according to the present invention. FIG. 17 compares the method according to the present invention for determining the solution search range based on the past positioning position and the position obtained by the Doppler shift, and the conventional method. As shown in FIG. 17, FIX solutions (ambiguities) are obtained at times T1 to T7. At times T1 to T7, the number of visible satellites is 5, 5, 4, 4, 3, 2, 4.

  As can be seen from the results of FIG. 17, at times T1 and T2, the ambiguity can be determined (the solution is FIX) not only by the method according to the present invention but also by the conventional method. However, at times T3 and T4, the error in the search center of the solution obtained by the pseudorange is large in the conventional method, so that the solution cannot be FIXed, whereas in the method according to the present invention, the solution is correctly corrected. Have been able to. When the number of visible satellites is reduced to 3 at time T5 and the number of visible satellites is 4 at time T7, according to the method of the present invention, the position at time T4 and the times T4 to T7 obtained by Doppler shift are used. By using the velocity vector at, the solution can be correctly FIXed even at time T7.

  FIG. 18 is an explanatory view showing another example of the positioning result according to the present invention. As shown in FIG. 18, in the present invention using a velocity vector, when the average number of visible satellites is 5.3, the rate at which a solution can be determined in one epoch is 80.9%. When the average number of visible satellites is 6.7, the rate at which a solution can be determined by one epoch is 88.4%. In each case, the rate at which a solution can be determined by one epoch compared to the conventional method. It can be seen that there is a significant improvement.

  As described above, according to the present invention, since the ambiguity is determined after limiting the search range of the solution from the past positioning position and the velocity vector, the ambiguity is calculated as compared with the conventional method. Speed up, especially improve the possibility of calculating the solution in one measurement (one epoch). Thereby, the time which can perform RTK-GPS positioning in an urban area etc. can be increased significantly.

  In the above-described embodiment, in order to obtain the position accurately even in urban areas where shielding of signals (carrier waves) from satellites frequently occurs, the solution is performed by one measurement (one epoch). The search range of the solution is limited by using a reliable previous positioning position and a velocity vector from the positioning position. However, the signal from the satellite is not limited to once but continuously. Can be obtained by an ordinary method using these time-series data, and a method using a velocity vector and an ordinary method can be appropriately combined.

  The disclosed embodiments are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

It is a schematic diagram which shows the outline | summary of an example of the positioning method using the positioning apparatus which concerns on this invention. It is a block diagram which shows an example of a structure of the positioning object containing the positioning apparatus which concerns on this invention. It is explanatory drawing which shows an example of the outline | summary of road shape information. It is explanatory drawing which shows the other example of the outline | summary of road shape information. It is explanatory drawing which shows an example of the data structure of road shape information. It is a flowchart which shows the procedure of the positioning process of a positioning apparatus. It is explanatory drawing which shows an example of the positioning by DGPS. It is a conceptual diagram which shows the example which restrict | limits the search range of a solution using road shape information. It is explanatory drawing which shows the example which calculates a velocity vector using a Doppler shift. It is explanatory drawing which shows an example of the method of determining the search range of a solution. It is explanatory drawing which shows typically a mode that the search range of a solution is limited. It is a flowchart which shows the procedure of the positioning process by the Hatch method of a positioning apparatus. It is a flowchart which shows the procedure of the positioning process by the Hatch method of a positioning apparatus. It is a flowchart which shows the procedure of the positioning process by the Hatch method of a positioning apparatus. It is a flowchart which shows the procedure of the positioning process by the Hatch method of a positioning apparatus. It is explanatory drawing which shows an example of the reliability of the initial value of ambiguity. It is explanatory drawing which shows an example of the positioning result by this invention. It is explanatory drawing which shows the other example of the positioning result by this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Positioning device 11 Processing unit 12 Satellite communication unit 121 Code phase measurement unit 122 Carrier phase measurement unit 123 Carrier frequency measurement unit 13 Communication unit 14 Storage unit 15 Map database 20 Autonomous sensor unit 201 Wheel speed sensor 202 Gyro sensor 100 Positioning target 300 Base station

Claims (8)

In a positioning device that includes a receiving unit that receives each carrier wave transmitted by a plurality of positioning satellites and performs positioning based on the phase of each carrier wave received by the receiving unit,
Position acquisition means for acquiring its own position before positioning;
Speed vector calculation means for calculating a speed vector from the position acquired by the position acquisition means;
Ambiguity determining means for determining the ambiguity included in the phase using the position acquired by the position acquiring means and the velocity vector calculated by the velocity vector calculating means;
A positioning device comprising: positioning means for positioning its own position based on the ambiguity determined by the ambiguity determining means. The velocity vector calculating means includes
The positioning device according to claim 1, wherein a velocity vector is calculated based on a Doppler shift of each carrier wave received by the receiving unit. Speed information acquisition means for acquiring information about speed;
An angle acquisition means for acquiring a turning angle;
The velocity vector calculating means includes
The positioning device according to claim 1, wherein a speed vector is calculated based on the speed acquired by the speed information acquiring unit and the turning angle acquired by the angle acquiring unit. A position reliability determination means for determining the reliability of the position acquired by the position acquisition means;
The ambiguity determining means is
The positioning device according to any one of claims 1 to 3, wherein ambiguity is determined in accordance with a determination result of the position reliability determination means. An initial value calculating means for calculating an initial value of ambiguity included in the phase of each carrier;
Initial value reliability determination means for determining the reliability of the initial value calculated by the initial value calculation means;
Search range setting means for setting the size of the search range of the ambiguity included in the phase according to the reliability determined by the initial value reliability determination means,
The ambiguity determining means is
The positioning device according to claim 1, wherein ambiguity is determined based on the search range set by the search range setting means. The search range setting means includes
6. The positioning device according to claim 5, wherein the size of the search range is set according to the passage of time from the time when the position before positioning is positioned. In a computer program for causing a computer to function as a means for determining an ambiguity included in the phase of each carrier wave transmitted by a plurality of positioning satellites,
Computer
Speed vector calculation means for calculating a speed vector from its own position before positioning;
A computer program that functions as ambiguity determining means for determining an ambiguity included in a phase using the position and the calculated velocity vector. In a positioning method by a positioning device that includes a receiving unit that receives each carrier transmitted by a plurality of positioning satellites, and performs positioning based on the phase of each carrier received by the receiving unit,
Get your own position before positioning,
Calculate the velocity vector from the acquired position,
Using the acquired position and the calculated velocity vector, determine the ambiguity included in the phase,
A positioning method characterized by positioning its own position based on the determined ambiguity.

Images