Positioning device and positioning method

Abstract

The present invention aims to provide a technique capable of appropriately utilizing centimeter-level positioning in the field of automobiles. A positioning device obtains a single positioning solution including a vehicle position, obtains a float solution including the vehicle position and a carrier phase bias, obtains an integer value bias, obtains a fixed solution including the vehicle position, sets any one of the single positioning solution, the float solution, the fixed solution, and a non-positioning solution indicating that a solution does not exist, as a positioning solution, and predicts a positioning error of the positioning solution as a positioning error of the vehicle position per epoch.

Description

Technical Field

[0001] This disclosure relates to positioning devices and positioning methods. Background Technology

[0002] In the field of surveying, three frequencies, including military L2P and civilian L1 and L5, are emitted from multiple GNSS satellites such as GPS, GLONASS, Galileo, Beidou, and QZSS. Centimeter-level positioning can be achieved in corresponding GNSS receivers. In lower-precision standalone positioning methods, the pseudo-range of the positioning signal, calculated based on the radio wave propagation time from the satellite to the vehicle, is used as the primary observation data. However, in centimeter-level positioning, carrier phase is also used as the primary observation data. Therefore, the method using carrier phase is called carrier phase positioning.

[0003] Carrier phase positioning methods include RTK (Real-Time Kinematic) and PPP-RTK (Precise Point Positioning RTK). In carrier phase positioning, the cumulative value is calculated by continuously measuring the carrier phase angle of the positioning signal demodulated by the GNSS receiver. However, if continuous observation is interrupted due to cycle slips or other reasons, this cumulative value is reset, and the carrier phase offset changes when the reset occurs. In other words, the carrier phase offset does not change during continuous reception of radio waves from multiple positioning satellites. Therefore, while it is possible to calculate the carrier phase offset with high precision in a single operation, it is not possible to calculate it for each epoch.

[0004] In carrier phase positioning methods that include individual positioning methods, a total of three solutions can be obtained as positioning solutions, namely: two solutions when the enhancement signal is used (floating-point solution) and one fixed solution (fixed solution), and one individual positioning solution when the enhancement signal is not used.

[0005] Centimeter-level positioning used in surveying is performed in an environment free of surrounding structures that could cause radio wave reception failures (this environment is known as open sky). The accuracy of a fixed solution is referred to as centimeter-level accuracy, with a fix rate expected to be at least 95%, typically above 99%. On the other hand, floating-point solutions are less accurate than fixed solutions (e.g., centimeter-to-meter accuracy), therefore, floating-point solutions are generally not used when centimeter-level positioning is the objective.

[0006] The aforementioned GNSS receivers can perform high-precision positioning, but they are very expensive and cannot be used in mass-produced vehicles. Positioning devices in the automotive field use inexpensive GNSS receivers and GNSS antennas to receive civilian L1 and L2C signals transmitted from multiple GNSS satellites such as GPS, GLONASS, Galileo, Beidou, and QZSS, and perform positioning calculations using carrier phase positioning. However, compared to expensive GNSS antennas used for measurement, the C / N value (carrier-to-noise ratio) of positioning signals received by inexpensive GNSS antennas is generally lower, even in open skies, and the average C / N sometimes varies depending on the frequency band of the L1 and L2C signals. Furthermore, various problems exist due to the rich variations in the radio wave reception environment around vehicles, such as from open skies to multipath environments of varying sizes, environments where some satellite radio waves above the vehicle are blocked, and tunnel environments where all satellite radio waves are blocked. To address these problems, technologies such as those in Patent Documents 1-4 have been proposed. Existing technical documents Patent documents

[0007] Patent Document 1: Japanese Patent No. 5590010 Patent Document 2: Japanese Patent No. 5083749 Patent Document 3: International Publication No. 2017 / 138502 Patent Document 4: Japanese Patent Application Publication No. 2010-071686 Summary of the Invention The technical problem that the invention aims to solve

[0008] In vehicles traveling on the road, sometimes a "miss fix" occurs, where the accuracy of the vehicle position obtained from a fixed solution drops to the meter level. Furthermore, if the fix rate of the fixed solution decreases, the floating rate of the floating-point solution with uncertain accuracy will actually increase. However, current technologies do not predict the positioning errors of both fixed and floating-point solutions in real time, thus presenting the problem that centimeter-level positioning is difficult to achieve in the automotive field.

[0009] Therefore, this disclosure was made in view of the above-mentioned problems, and its purpose is to provide a technology that enables centimeter-level positioning in the automotive field. Technical means for solving technical problems

[0010] The positioning apparatus disclosed herein includes: a GNSS acquisition unit that acquires observation data including pseudorange, carrier phase, and Doppler offset frequency, and orbit data of multiple GNSS satellites for each positioning signal from multiple GNSS satellites; a positioning enhancement data acquisition unit that acquires positioning enhancement data from positioning enhancement satellites or the Internet; a positioning satellite selection unit that selects a positioning satellite from multiple GNSS satellites; and a separate positioning solution calculation unit that calculates the individual positioning solution based on the observation data and orbit data of the positioning satellite without using positioning enhancement data. The system includes: a positioning solution; a floating-point solution calculation unit that calculates a floating-point solution including carrier phase offset based on satellite observation data, orbit data, and positioning augmentation data; a search and verification unit that calculates an integer offset based on the carrier phase offset of the floating-point solution; a fixed solution calculation unit that calculates a fixed solution based on satellite observation data, orbit data, positioning augmentation data, and integer offset; and a satellite positioning error prediction unit that sets any one of the following as the positioning solution: a standalone positioning solution, a floating-point solution, a fixed solution, and a non-positioning solution indicating that the solution does not exist, and predicts the positioning error of the positioning solution at each epoch. Invention Effects

[0011] According to this disclosure, the occurrence of erroneous fixation can be suppressed, and the positioning error of the positioning solution can be predicted in real time. Therefore, centimeter-level positioning can be used in the automotive field.

[0012] The purpose, features, aspects, and advantages of this disclosure will become more apparent from the following detailed description and accompanying drawings. Attached Figure Description

[0013] Figure 1 This is a block diagram showing the structure of the positioning device according to Embodiment 1. Figure 2 This is a flowchart illustrating the operation of the positioning device according to Embodiment 1. Figure 3 This is a diagram illustrating an example of the results of calculating the residuals of the pseudo-distance involved in Implementation 1. Figure 4 This is a schematic diagram illustrating the single and double differences of the observation data involved in Implementation Method 1. Figure 5 This is a diagram illustrating an example of the relationship between the prediction error and the actual error of the floating-point solution involved in Implementation 1. Figure 6 This is a graph illustrating an example of the relationship between the prediction error and the actual error of the fixed solution involved in Implementation 1. Figure 7 This is a diagram illustrating the re-search of ambiguity involved in Variation 4 of Implementation 1. Figure 8 This is a block diagram showing the structure of the positioning device according to Embodiment 2. Figure 9 This is a block diagram showing the structure of the positioning device involved in a variation of Embodiment 2. Figure 10 This is a block diagram showing the structure of the positioning device according to Embodiment 3. Figure 11 This is a flowchart illustrating the operation of the positioning device according to Embodiment 3. Figure 12 This is a flowchart illustrating the operation of the positioning device according to Embodiment 3. Figure 13 This is a diagram used to illustrate the operation of the positioning device involved in Embodiment 3. Figure 14 This is a diagram illustrating the operation of the positioning device involved in Variation 2 of Embodiment 3. Figure 15 This is a diagram illustrating the operation of the positioning device involved in Variation 2 of Embodiment 3. Figure 16 This is a diagram illustrating the operation of the positioning device involved in Variation 2 of Embodiment 3. Figure 17 This is a block diagram illustrating the structure of the driving assistance system according to Embodiment 4. Figure 18 This is a diagram showing an example of a display screen according to Embodiment 4. Figure 19 This is a diagram showing an example of a display screen according to Embodiment 4. Figure 20 This is a diagram showing an example of a display screen according to Embodiment 4. Figure 21 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 22 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 23 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 24 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 25 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 26 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 27 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 28 This is a diagram illustrating an example of a display screen involved in a variation of embodiment 4. Figure 29 This is a block diagram illustrating the structure of the driving assistance system according to Embodiment 5. Figure 30 This is a diagram used to illustrate the stereo camera and millimeter-wave radar involved in Embodiment 5. Figure 31 This is a diagram used to illustrate the stereo camera and millimeter-wave radar involved in Embodiment 5. Figure 32 This is a diagram used to illustrate a monitoring example performed by the stereo camera and millimeter-wave radar according to Embodiment 5. Figure 33 This is a diagram used to illustrate a monitoring example performed by the stereo camera and millimeter-wave radar according to Embodiment 5. Figure 34 This is a diagram used to illustrate a monitoring example performed by the stereo camera and millimeter-wave radar according to Embodiment 5. Figure 35 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 36 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 37 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 38 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 39 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 40 This is a diagram illustrating a monitoring example performed by a stereo camera and millimeter-wave radar in a variation of Embodiment 5. Figure 41 This is a block diagram illustrating the structure of the driving assistance system according to Embodiment 6. Figure 42 This is a diagram used to illustrate the road condition sensor and laser vehicle height gauge involved in Embodiment 6. Figure 43 This is a block diagram showing the structure of the driving assistance system according to Embodiment 7. Figure 44 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 45 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 46 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 47 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 48 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 49 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 50 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 51 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 52 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 53 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 54 This is a diagram illustrating an example of vehicle control in the driving assistance system according to Embodiment 7. Figure 55 This is a block diagram illustrating the hardware structure of the positioning device involved in other variations. Figure 56 This is a block diagram illustrating the hardware structure of the positioning device involved in other variations. Detailed Implementation

[0014] <Implementation Method 1> Figure 1 This is a block diagram showing the structure of the positioning device according to Embodiment 1. Hereinafter, the vehicle equipped with the positioning device and which is of interest will sometimes be referred to as "this vehicle", its location as "this vehicle location", the positioning device as "mobile station", and the GNSS satellite as "satellite".

[0015] Figure 1The positioning device is a GNSS acquisition unit, which includes a GNSS receiver 11 containing a GNSS antenna, a positioning enhancement signal receiver 12 as a positioning enhancement data acquisition unit, and a satellite positioning unit 13.

[0016] The GNSS receiver 11 receives positioning signals from radio waves transmitted in a specified frequency band from multiple GNSS satellites, including GPS satellites, located above the vehicle. In this embodiment 1, the positioning signal is a dual-frequency signal, comprising a first positioning signal, namely the L1 signal, and a second positioning signal, namely the L2C signal, with a frequency band different from the L1 signal, but is not limited to this. Based on the positioning signals broadcast by the multiple GNSS satellites, the GNSS receiver 11 generates time data, observation data, and orbit data.

[0017] Timing data includes, for example, the timing used for synchronization. For each positioning signal, the observation data includes the rover's pseudorange, carrier phase, Doppler offset frequency, and ionospheric delay error. Orbit data is required for position calculations from multiple GNSS satellites and includes broadcast history.

[0018] The positioning enhancement signal receiver 12 is connected to the server (not shown) of the positioning enhancement signal distribution provider via Internet line 12a. Furthermore, authentication of the access point is performed during connection. The positioning enhancement signal receiver 12 receives positioning enhancement signals, including positioning enhancement data corresponding to the carrier phase positioning method, from the connected server at predetermined intervals by appropriately transmitting the location data of the mobile station corresponding to the carrier phase positioning method. The positioning enhancement data includes the location data and observation data of a reference station serving as a reference point. The reference station and observation data can be virtual, while the location data of the reference station can be fixed data. In this embodiment 1, the carrier phase positioning method is described as a VRS (Virtual Reference Station) method with a virtual reference point, but it could also be an RTK method.

[0019] The satellite positioning unit 13 includes a satellite selection unit 131 for positioning, a separate positioning solution calculation unit 132, a floating-point solution calculation unit 133, an ambiguity search and verification unit 134, a fixed solution calculation unit 135, and a satellite positioning error prediction unit 136.

[0020] The positioning satellite selection unit 131 selects a positioning satellite from multiple GNSS satellites. The time data, observation data, and orbit data of the selected positioning satellite are used by separate positioning solution calculation units 132, floating-point solution calculation units 133, and fixed-point solution calculation units 135.

[0021] The standalone positioning solution calculation unit 132 calculates the standalone positioning solution, including the built-in clock error of the GNSS receiver, based on the time data, observation data and orbit data of the positioning satellite, without using positioning augmentation data, according to the standalone positioning method.

[0022] The floating-point solution calculation unit 133 calculates the floating-point solution and carrier phase offset based on the carrier phase positioning method of VRS (Virtual Reality System), using satellite time data, observation data, orbit data, and virtual observation data, including the coordinates of a virtual reference point virtually set near the vehicle's position by the aforementioned server and the virtual observation results at those coordinates. In this embodiment 1, as described above, the carrier phase positioning method is VRS. The floating-point solution is a real value.

[0023] The ambiguity search and verification unit 134 searches and detects according to the LAMBDA (Least-square ambiguity decorrelation adjustment method) method described later, and calculates the integer value of the carrier phase offset, which is called ambiguity, based on the carrier phase offset represented by the real value of the floating-point solution.

[0024] The fixed solution calculation unit 135 calculates the fixed solution according to the carrier phase positioning method of VRS, based on the time data, observation data and orbit data of the positioning satellite, the coordinates of the virtual reference point virtually set by the server near the vehicle's position and the virtual observation results under the coordinates, and the integer value offset.

[0025] The satellite positioning error prediction unit 136 sets one of the following as the positioning solution for satellite positioning: a standalone positioning solution, a floating-point solution, a fixed solution, and a non-positioning solution indicating that the solution does not exist. Furthermore, the satellite positioning error prediction unit 136 predicts the positioning error of the positioning solution for each epoch.

[0026] <Action> Next, the operation of the positioning device according to Embodiment 1 will be explained. Figure 2 This is a flowchart illustrating the actions of the positioning device for each epoch. Figure 2 The operation is performed by the satellite positioning unit 13. The following mainly describes the VRS method, but the RTK method is described in the same way.

[0027] First of all, Figure 2 In step S201, the processing of the positioning device is initialized.

[0028] In step S202, the satellite positioning unit 13 determines whether there are 4 or more GNSS satellites (hereinafter also referred to as "receiving satellites") that can receive positioning signals from the GNSS receiver 11. If there are fewer than 4 receiving satellites, the process proceeds to step S211; if there are 4 or more receiving satellites, the process proceeds to step S203.

[0029] In step S203, the positioning satellite selection unit 131 selects a positioning satellite from multiple GNSS satellites. The selection of the positioning satellite will be explained below.

[0030] Figure 3 This is a diagram illustrating an example of the results of calculating the residuals of pseudo-range as one of the observation data. Currently, GNSS receivers that output pseudo-range residuals are known, but as... Figure 3 (c) In this way, the pseudo-distance residual only indicates the occurrence of multipath.

[0031] Therefore, in this embodiment 1, the positioning satellite selection unit 131 determines the positioning signal based on a comparison between the pseudo-range contained in the received positioning signal and the carrier phase or Doppler offset frequency. Figure 3 (a) The first pseudo-distance residual. Furthermore, the calculation of the first pseudo-distance residual can, for example, be performed using the calculation methods described in Japanese Patent No. 4988028 or Japanese Patent No. 6482720.

[0032] With both L1 and L2C signals available (i.e., both signals are received), the positioning satellite selection unit 131 uses ionospheric-free linear coupling to cancel the ionospheric delay errors of both the L1 and L2C signals. By canceling these ionospheric delay errors, the positioning satellite selection unit 131 determines... Figure 3 (b) such a second pseudo-distance residual.

[0033] Next, the positioning satellite selection unit 131 performs an initial selection based on at least one of the first pseudo-range residual and the second pseudo-range residual, as well as the elevation angle and C / N value, selecting GNSS satellites with communication quality above the first threshold required to achieve positioning accuracy as temporary positioning satellites.

[0034] Then, if the amount of observation data from the temporary positioning satellite is below a predetermined amount, the positioning satellite selection unit 131 selects a temporary positioning satellite as the positioning satellite. On the other hand, if the amount of observation data from the temporary positioning satellite exceeds a predetermined amount, the positioning satellite selection unit 131 performs a secondary selection based on the communication quality of the temporary positioning satellite, selecting a positioning satellite from the temporary positioning satellites whose amount of observation data is below a threshold. For example, the positioning satellite selection unit 131 performs the secondary selection based on at least one of the first and second pseudo-range residuals of the temporary positioning satellite, as well as the elevation angle and C / N value, selecting a GNSS satellite with a communication quality higher than a second threshold that is higher than the first threshold required to achieve positioning accuracy as the temporary positioning satellite. The pseudo-range residuals used in the initial selection and the pseudo-range residuals used in the secondary selection can be the same or different.

[0035] Furthermore, among the positioning satellites used by the independent positioning solution calculation unit 132, GNSS satellites with only single-frequency positioning signals can be selected. On the other hand, among the positioning satellites used by the floating-point solution calculation unit 133 and the fixed-point solution calculation unit 135, GNSS satellites with dual-frequency positioning signals are preferentially selected compared to GNSS satellites with only single-frequency positioning signals. Based on this selection, the accuracy of both floating-point and fixed solutions is improved.

[0036] In positioning satellites, those with higher C / N ratios, higher elevation angles, and higher pseudorange quality are designated as primary satellites, while other positioning satellites are designated as secondary satellites. The following is an example describing the selection criteria for positioning satellites.

[0037] <Selection Criteria for Using Satellites for Positioning> (1) The pseudo-distance residuals (mainly the estimated multipath error) of L1 and L2C signals are below the specified values.

[0038] (2) The elevation angle of the GNSS satellite is above the specified value.

[0039] (3) The state in which the C / N values ​​of L1 signal and L2C signal are above the specified value continues for more than the specified epoch.

[0040] (4) The carrier phases of the L1 and L2C signals are observed after a specified time following an unobserved cycle slip.

[0041] (5) The observation data of the GNSS satellites that are common between the virtual reference point and the rover station are complete (this selection condition is only applicable to the case where the carrier phase positioning method is VRS).

[0042] (6) The delay time of the observation data of the virtual reference point is less than the specified time.

[0043] (7) The number of GNSS satellites with both L1 and L2C signals is above the specified number (the reason is that in order to eliminate the influence of ionospheric delay error in calculating pseudo-range residuals, GNSS satellites with both frequencies are better).

[0044] Furthermore, when the observation data only includes carrier phase, the number of simultaneous equations needed to solve for the unknowns is insufficient. In contrast, in this embodiment 1, the observation data includes not only carrier phase but also pseudo-range, thus increasing the number of simultaneous equations needed to solve for the unknowns.

[0045] In step S204, the individual positioning solution calculation unit 132 calculates the individual positioning solution, which includes the vehicle's position and the built-in clock error of the GNSS receiver, based on the pseudo-range contained in the observation data of more than four GNSS satellites and the broadcast history contained in the orbit data, according to the individual positioning method.

[0046] In step S205, the satellite positioning unit 13 determines whether there are 5 or more satellites used for positioning and whether the number of simultaneous equations is greater than the number of unknowns. If there are 5 or more satellites used for positioning and the number of simultaneous equations is greater than the number of unknowns, the process proceeds to step S206. If there are fewer than 5 satellites used for positioning and the number of simultaneous equations is less than the number of unknowns, the process proceeds to step S211.

[0047] In step S206, the floating-point solution calculation unit 133 performs convergence calculations within an epoch according to steps 1 to 4 of the VRS method as follows, to obtain the floating-point solution and carrier phase offset as unknowns. Furthermore, the carrier phase offset obtained here is a real number, not an approximate integer.

[0048] <Step 1> The floating-point solution calculation unit 133 obtains the decision results of the primary satellite and the secondary satellite of the positioning satellite selection unit 131.

[0049] <Step 2> The floating-point solution calculation unit 133 performs the following equations (1) to (8) used in the VRS mode for the carrier phase and simulated distance contained in the satellite observation data used for positioning.

[0050] [Mathematical Expression 1]

[0051] Equation (1) is the observation equation for the carrier phase of the reference station A and the main satellite. φ is the carrier phase [cycle], r is the geometric distance between the satellite and the reference station A [m], c is the speed of light [m / s], δt is the error of the built-in clock of the GNSS receiver [s], Δt is the error of the clock on the satellite [s], I is the ionospheric delay error [m], T is the tropospheric delay error [m], λ is the wavelength [m / cycle], N is the carrier phase offset, and ε φ It is the observation error of the carrier phase [m (meter)].

[0052] The "1" appended to the upper right of φ indicates n=1, representing data from the primary satellite. Additionally, the "n" appended to the upper right of φ in the following formulas indicates n=2, 3, ..., representing data from a secondary satellite. The "A" appended to the lower right of φ indicates data from base station A. Furthermore, the "B" appended to the lower right of φ in the following formulas indicates data from rover station B. The position of base station A is known, while the position of rover station B is unknown. For convenience, in the following explanations, base station A and rover station B will be collectively referred to as the receiver.

[0053] The difference between the observation equation of the reference station A in equation (1) and the observation equation of the mobile station B, which is expressed in the same way as equation (1), is used to approximate the ionospheric delay error and tropospheric delay error of the mobile station B with the same ionospheric delay error and tropospheric delay error as the reference station A, thus obtaining the following equations (2) and (3).

[0054] [Mathematical Expression 2]

[0055] [Mathematical Expression 3]

[0056] The "BA" appended to the lower right of φ indicates that the data is obtained by subtracting the data of the base station A from the data of the rover B. For example, φ 1 BA It is φ 1 B -φ 1 AThe φ with "BA" is called the receiver-to-receiver single difference observation, representing the difference between the base station and the mobile station for the same satellite, i.e., the path difference. That is, equation (2) represents the difference between the base station and the mobile station for the main satellite (n=1), i.e., the path difference, and represents the receiver-to-receiver single difference for the carrier phase of the main satellite (n=1). Equation (3) represents the difference between the base station and the mobile station for the secondary satellites (n=2, 3, ...), i.e., the path difference, and represents the receiver-to-receiver single difference for the carrier phase of the secondary satellites (n=2, 3, ...). The difference between equation (2) and equation (3) yields equation (4).

[0057] [Mathematical Expression 4] in, r n B ={(x B -x n ) 2 +(y B -y n ) 2 +(z B -z n ) 2} 1 / 2 (x B y B , z B Location of the rover's GNSS receiver (x n y n , z n ): Position of satellite n

[0058] The "1n" appended to the upper right of φ indicates the inter-satellite difference data, obtained by subtracting the inter-satellite difference data from the inter-receiver difference data of satellites (n=2, 3, ...) from the inter-receiver difference data of the primary satellite (n=1). For example, φ 1n BA It is φ 1 BA -φ n BA φ 1n BA Also known as double-difference observations.

[0059] Figure 4 This is a schematic diagram illustrating the single and double differences of the observation data used by the satellite positioning unit 13 for positioning. (Example) Figure 4As shown, the VRS method is a relative positioning method that determines the relative position of the rover B with respect to a base station A whose coordinates are known. Therefore, the floating-point solution calculation unit 133 calculates the difference between the distance between the satellite and the base station A and the distance between the satellite and the rover B, i.e., the path difference, for each satellite, as the inter-receiver single difference SDR. Then, the floating-point solution calculation unit 133 calculates the difference between the inter-receiver single difference SDR of the master satellite S1 and the inter-receiver single difference SDR of the slave satellite Sn, as the double difference observation value.

[0060] Equations (1) to (4) above are about the carrier phase. Equations (5) to (8) below, which are the same as those above (1) to (4), also hold true for pseudo-range.

[0061] [Mathematical Expression 5] ρ 1 A =r 1 A +C(δt A -Δt 1 )+I 1 A +T 1 A +ε ρA [m]…(5)

[0062] [Mathematical Expression 6] ρ 1 BA =r 1 BA +Cδt BA +ε ρBA …(6)

[0063] [Mathematical Expression 7] ρ n BA =r n BA +Cδt BA +ε ρBA …(7)

[0064] [Mathematical Expression 8] ρ 1n BA +r 1n A =r 1n B …(8)

[0065] Equation (5) is the observation equation for the pseudo-range between the base station A and the master satellite. ρ is the pseudo-range [m]. ερ is the observation error of the pseudo-range [m], also known as the pseudo-range residual. Equations (6) and (7) are obtained by taking the difference between the observation equation of the base station A in equation (5) and the observation equation of the rover B, which is expressed in the same way as in equation (5). Equation (8) is obtained from the difference between equations (6) and (7).

[0066] In equation (4) for carrier phase, the left side is the known number and the right side is the unknown number, which is the same as the number of satellites (i.e., n = 2, 3, ...). Similarly, in equation (8) for pseudo-range, the left side is the known number and the right side is the unknown number, which is the same as the number of satellites (i.e., n = 2, 3, ...).

[0067] <Step 3> The floating-point solution calculation unit 133 calculates the observation error of the single difference by comparing the difference between the observed and predicted values ​​of the single difference, i.e., the residual, and the difference between the range of the increment (delta) and the rate of change of distance. The calculation of the observation error of the single difference can, for example, use the calculation methods described in Japanese Patent No. 4988028, Japanese Patent No. 5855249, or Japanese Patent No. 6482720.

[0068] <Step 4> The floating-point solution calculation unit 133 uses the extended Kalman filter shown in equations (9) to (13) to perform convergence calculation in order to obtain the floating-point solution including the estimated value of the vehicle position, the carrier phase offset and the observation error covariance matrix.

[0069] [Mathematical Expression 9]

[0070] [Mathematical Expression 10]

[0071] [Mathematical Expression 11]

[0072] [Mathematical Expression 12]

[0073] [Mathematical Expression 13]

[0074] Equation (9) represents the unknowns corresponding to the floating-point solutions on the right-hand side of equations (4) and (8), which are equivalent to the state variables of the extended Kalman filter. Additionally, in equation (9), i = 1 represents the L1 signal, and i = 2 represents the L2C signal. Equation (10) represents the knowns corresponding to the floating-point solutions on the left-hand side of equations (4) and (8), which are equivalent to the observations of the extended Kalman filter.

[0075] Equation (11) is equivalent to the propagation step (also known as the prediction step) of the extended Kalman filter, which is a formula for predicting the state from a certain time to the next time by linear approximation. Additionally, Q... k The element is the observation error of the single difference obtained in step 3. Equation (12) is equivalent to the update step (also called the filtering step) of the extended Kalman filter, which is used to update the current state based on the comparison between the predicted state and the observation. Equation (13) is the expression for the observation error covariance matrix shown in equation (12).

[0076] Furthermore, the floating-point calculation unit 133 is configured to perform calculations under predetermined conditions (e.g., ...). Figure 2 Under the conditions of steps S202 and S205, a floating-point solution is obtained for each epoch, regardless of whether the ambiguity search verification unit 134 obtains an integer bias.

[0077] In step S207, the ambiguity search and verification unit 134 determines whether it is necessary to search for the integer value offset of the positioning satellite, i.e., the ambiguity. In this embodiment 1, if either (1) the L1 and L2C used by the positioning satellite are cycle-slipped, or (2) the integer value offset of the L1 and L2C used by the positioning satellite is not calculated, the ambiguity search and verification unit 134 determines that it is necessary to search for the integer value offset. If it is determined that it is necessary to search for the integer value offset, the process proceeds to step S208; if it is determined that it is not necessary to search for the integer value offset, the process proceeds to step S210.

[0078] In step S208, the ambiguity search and verification unit 134 searches for the integer value offset of the carrier phase, i.e., the candidate point of ambiguity, based on the carrier phase offset represented by the real value of the floating-point solution.

[0079] Here, if the carrier phase offset represented by a real value is simply rounded to obtain an integer offset value, it will become an incorrect value due to the strong correlation shown in the search space, resulting in a decrease in positioning accuracy. Therefore, the ambiguity search and verification unit 134 uses the LAMBDA method to search for the integer offset value according to steps 1 to 4.

[0080] <Step 1> The ambiguity search and verification unit 134 uses the carrier phase offset and covariance matrix represented by the real values ​​of the floating-point solution to perform offset decorrelation as much as possible. For example, for the error covariance matrix of the extended Kalman filter, the ambiguity search and verification unit 134 performs decorrelation by performing diagonalization based on orthogonal matrices as much as possible.

[0081] <Step 2> The ambiguity search verification unit 134 repeats the LDL test on the results obtained in step 1. T Decomposition and UDU T Decompose the z-transformation matrix.

[0082] <Step 3> The ambiguity search verification unit 134 defines a search space (e.g., an elliptic) for integer value bias based on the result obtained in step 2, and searches for multiple candidate points for integer value bias contained in the search space.

[0083] <Step 4> The fuzziness search and verification unit 134 defines a narrower new search space based on the multiple candidate points obtained in step 3, and searches for multiple candidate points with integer bias within this search space. The fuzziness search and verification unit 134 repeats step 4 until there are two candidate points with integer bias. Additionally, before there are two candidate points with integer bias, processing time of approximately several epochs may be required.

[0084] In step S209, the fuzziness search and verification unit 134 verifies the candidate points for integer value bias found in step S208. In this embodiment 1, the fuzziness search and verification unit 134 calculates the ratio of the sum of squares of the residuals of the first candidate to the sum of squares of the residuals of the second candidate. The numerator of this ratio is the second candidate, and the denominator is the first candidate. If the ratio is above a threshold, the fuzziness search and verification unit 134 determines that the verification is successful, sets the first candidate as the integer value bias, and proceeds to step S210. If the ratio is below the threshold, the fuzziness search and verification unit 134 determines that the verification is unsuccessful, and proceeds to step S211.

[0085] In step S210, the fixed solution calculation unit 135 uses the qualified integer bias and the extended Kalman filter shown in equations (9) to (13) to perform convergence calculation within the epoch to obtain the fixed solution.

[0086] In step S211, the satellite positioning error prediction unit 136 sets any one of the following in the latest epoch: a single positioning solution, a floating-point solution, a fixed solution, or a non-positioning solution indicating that the solution does not exist, as a positioning solution state. In the positioning solution state, the flag indicating the presence or absence of each positioning solution is allocated using bits. Sometimes multiple positioning solutions exist simultaneously, and sometimes no positioning solution exists (i.e., a non-positioning solution).

[0087] The satellite positioning error prediction unit 136 predicts the positioning error of the floating-point solution according to the first calculation formula, based on observation data and internal data of the positioning operation that are correlated with the actual error of the floating-point solution. The first calculation formula is a formula that expresses the relationship between the observation data, the internal data, and the positioning error of the floating-point solution. The internal data includes, for example, information on at least one of the following: the number of satellites with only single frequency and satellites with dual frequency; the configuration of the positioning satellites; the observation status of the positioning satellites; the vehicle speed; the time elapsed since receiving the positioning enhancement data; the observation error of the extended Kalman filter; the error covariance of the extended Kalman filter; and the convergence state of the floating-point solution.

[0088] Furthermore, the satellite positioning error prediction unit 136 learns the first calculation formula through machine learning and other means, based on observation data, internal data, and the actual error of the floating-point solution calculated using electronic reference points in subsequent processing. Additionally, the parameters of the first calculation formula may include not only observation data and internal data, but also the actual error of the floating-point solution and the residuals of the observation data, or may only include internal data. Figure 5 In the section on floating-point solutions, an example is shown illustrating the relationship between the error predicted by the satellite positioning error prediction unit 136 (i.e., the predicted error) and the actual error. Additionally, Figure 5 The straight line in the figure shows the ideal relationship between the prediction error and the actual error.

[0089] Similarly, the satellite positioning error prediction unit 136 predicts the positioning error of the fixed solution based on observation data and internal data of the positioning operation that are correlated with the actual error of the fixed solution, according to the second calculation formula. The second calculation formula is a formula that expresses the relationship between the observation data, the internal data, and the positioning error of the fixed solution. The internal data can be the same as the internal data mentioned above.

[0090] Furthermore, the satellite positioning error prediction unit 136 learns the second calculation formula based on observation data, internal data, and the actual error of the fixed solution calculated using electronic reference points in subsequent processing, etc., through machine learning and other means. Additionally, the parameters of the second calculation formula may include not only observation data and internal data, but also the actual error of the fixed solution and the residuals of the observation data, or may only include internal data. Figure 6 In the example given regarding the fixed solution, the relationship between the error predicted by the satellite positioning error prediction unit 136 (i.e., the predicted error) and the actual error is shown. Additionally, Figure 6 The straight line in the figure shows the ideal relationship between the prediction error and the actual error.

[0091] As described above, in step S211, the satellite positioning error prediction unit 136 calculates the positioning error for each epoch using both floating-point and fixed-point solutions. After step S211, the process returns to step S201.

[0092] <Summary of Implementation Method 1> According to the positioning device described in this embodiment, the positioning error of the positioning solution can be predicted for each epoch. Therefore, it is possible to appropriately detect errors in the fixed solution where the vehicle's position accuracy drops to the meter level, and the fixed solution can be utilized based on the suspected intensity of the accuracy drop. Similarly, for floating-point solutions, a floating-point solution matching the desired positioning accuracy can be selected and utilized, thus improving the utilization of both floating-point and fixed solutions. As a result, centimeter-level positioning can be utilized in the automotive field.

[0093] Furthermore, in this embodiment 1, an initial selection of temporary positioning satellites with communication quality above a first threshold is performed based on at least one of the first pseudo-range residual and the second pseudo-range residual. If the amount of observation data from the temporary positioning satellites exceeds the threshold, a secondary selection of positioning satellites with communication quality above a second threshold, which is higher than the first threshold, is performed. Thus, when dual-frequency positioning signals are available, the multipath effect of pseudo-range can be accurately calculated by eliminating the influence of ionospheric delay errors contained in the pseudo-range residuals. As a result, pseudo-range residuals affected by multipath can be calculated, thus improving the accuracy of floating-point and fixed solutions. Furthermore, the multipath effect of pseudo-range can be accurately calculated for older generation GPS satellites with only L1 signals and GNSS satellites other than GPS (e.g., Galileo satellites) that do not use positioning enhancement signals. As a result, pseudo-range residuals affected by multipath can be calculated, thus improving the positioning rate and utilization of floating-point and fixed solutions in locations with fewer positioning satellites. Furthermore, even in open-sky environments where high-quality positioning signals exceeding the number of observations required to achieve the desired accuracy can be received, the system can prioritize higher-quality pseudoranges up to the upper limit of the observation data count. Therefore, it can suppress CPU utilization and computation time during positioning.

[0094] Furthermore, in this embodiment 1, a first calculation formula is learned, which represents the relationship between information such as the configuration of positioning satellites, the time elapsed since the acquisition of positioning augmentation data, the observation status of positioning satellites, and the convergence status of the floating-point solution, and the positioning error of the floating-point solution. A second calculation formula is also learned, which represents the relationship between this information and the positioning error of the fixed solution. This prevents the prediction error from becoming too small or too large compared to the actual error, thus improving the reliability of the prediction error and the usability of both the floating-point and fixed solutions.

[0095] <Variation Example 1> Figure 2In step S203, the positioning uses satellite selection unit 131 to calculate the first pseudo-range residual based on the received positioning signal, and calculates the second pseudo-range residual when both the dual-frequency L1 signal and L2C signal are available. However, the pseudo-range residual can be calculated using methods other than those described above.

[0096] <Variation Example 2> Figure 2 In step S211, the satellite positioning error prediction unit 136 learns the first calculation formula and the second calculation formula, but is not limited thereto. For example, the satellite positioning error prediction unit 136 can receive positioning results such as a fixed solution obtained by an external positioning device capable of measuring the actual error in real time with high precision (e.g., centimeter level), and adjust the first calculation formula and the second calculation formula based on prediction error calculation data including observation data and internal data of positioning calculation that are correlated with the actual error and the received positioning results. According to this structure, a prediction error that better matches the actual error can be calculated in vehicle-mounted environments such as roads where the vehicle travels, thus improving the reliability of the prediction error.

[0097] <Variation Example 3> Figure 2 In step S211, the satellite positioning error prediction unit 136 predicts the positioning error of the floating-point solution and the fixed solution based on the observation data and internal data of the positioning operation that are related to the actual error, according to the first calculation formula and the second calculation formula, but is not limited to this.

[0098] For example, in vehicle-mounted environments where the satellite radio wave reception environment may vary significantly due to the influence of surrounding structures, trees, etc., it is sometimes difficult to predict a positioning error at the same level as the actual error using the aforementioned prediction. Therefore, in such environments, the satellite positioning error prediction unit 136 can adjust the first and second calculation formulas with safety as a priority, so that the predicted error is larger than the actual error. However, if the prediction error is unnecessarily increased compared to the actual error, the usability of the fixed solution decreases. Therefore, in order to prevent the predicted error from becoming unnecessarily large compared to the actual error, it is preferable to adjust the first and second calculation formulas.

[0099] Furthermore, the data and calculation methods used in the first and second calculation formulas are not limited to those described above. In particular, Figure 5 The actual error range of the floating-point solution is greater than Figure 6 The actual error range of the fixed solution is wider. Therefore, if the first calculation formula is adjusted to make the prediction error of the floating-point solution appropriate, the target accuracy of the positioning device can be determined by considering both usability and safety.

[0100] <Variation Example 4> Figure 7 This is a graph showing the re-search of ambiguity (i.e., integer value bias) when the error fixation occurs. Figure 7 The diagram shows the vehicle 1, the fixed solution 2, the prediction error of the fixed solution 3, and the driving path of vehicle 1 5. Figure 7 The diagram shows the case where the larger the circle representing the prediction error 3 of the fixed solution, the greater the prediction error. When the vehicle 1, traveling on the road below the elevated road, passes through the intersection 4 with the elevated road, the prediction error 3 of the fixed solution increases.

[0101] Here, in Implementation 1, if the ambiguity search and verification unit 134 determines that the integer value offset verification is qualified, it will not calculate a new integer value offset during the unsuitable period. Furthermore, the unsuitable period is the period from when the integer value offset is calculated to when the radio waves of the positioning satellite are cut off or blocked, or the period from when the integer value offset is calculated to when the combination of positioning satellites is updated. Therefore, even if an ambiguity occurs... Figure 7 (a) Such an error is fixed and it is impossible to find a new integer bias value.

[0102] Therefore, the ambiguity search and verification unit 134 can determine the integer bias during the unsuitable period, provided that the positioning error of the fixed solution is greater than the first threshold and the pseudo-range residual of the satellite used for positioning is smaller than the second threshold. Based on this structure, as... Figure 7 As shown in (b), the fixed solution can be restored to the normal solution earlier and more reliably.

[0103] <Implementation Method 2> The positioning device described in Embodiment 2 uses a carrier phase positioning method, such as PPP-RTK, instead of the carrier phase positioning method in Embodiment 1. It can perform RTK positioning for measurement using only a rover station without using observation data from a base station. Furthermore, in Japan, a positioning enhancement service for PPP-RTK positioning is provided free of charge through a national infrastructure called CLAS (Centimeter Level Augmentation Service).

[0104] In PPP-RTK, satellite positioning errors are categorized into satellite orbit errors, satellite clock errors, satellite signal offset errors, ionospheric delay errors, tropospheric delay errors, receiver clock errors, and multipath errors. The global errors in satellite positioning (e.g., satellite orbit errors, satellite clock errors, satellite signal offset errors, etc.) are then presented more accurately, while local errors (e.g., ionospheric delay errors, tropospheric delay errors, etc.) are distributed as positioning enhancement signals by positioning enhancement satellites, i.e., quasi-zenith satellites. Based on this PPP-RTK method, high-precision positioning with centimeter-level accuracy can be achieved in open-sky environments.

[0105] Furthermore, the data communication capacity of the Quasi-Zenith Satellite (QZS) is limited. Therefore, the positioning augmentation data is lengthened and compressed to approximately 60 km square by the electronic reference point network (approximately 20 km square) of the Geospatial Information Authority of Japan. However, in order to achieve centimeter-level accuracy, the positioning augmentation data is lengthened based on dynamics with global errors as a common term, and local errors are reduced in dimension by modeling their spatial distribution and then compressed.

[0106] Figure 8 This is a block diagram showing the structure of the positioning device according to Embodiment 2 using CLAS. Hereinafter, structural elements that are the same as or similar to the above-described structural elements will be labeled with the same or similar reference numerals, and the different structural elements will be mainly described.

[0107] In this embodiment 2, the operation of the positioning enhancement signal receiver 12, the floating-point solution calculation unit 133, and the fixed solution calculation unit 135 is different from that in embodiment 1.

[0108] Positioning augmentation signals in PPP-RTK mode are defined by the Compact-SSR standard. Positioning augmentation data is divided into specified messages and broadcast from a quasi-zenith satellite as positioning augmentation signals such as L6 signals. Additionally, the positioning augmentation data includes satellite orbital errors, satellite clock errors, satellite code offsets, satellite phase offsets, satellite code-phase offsets, STEC correction data, grid correction data, tropospheric correction data, and related data, which are the main causes of positioning errors.

[0109] Unlike embodiment 1, the positioning enhancement signal receiver 12 receives positioning enhancement signals transmitted by the positioning enhancement satellite (i.e., the quasi-zenith satellite) at a predetermined period and acquires positioning enhancement data. The floating-point solution calculation unit 133, using the PPP-RTK carrier phase positioning method, calculates a floating-point solution including the vehicle's position and carrier phase offset based on the time data, observation data, and orbit data of the positioning satellite, as well as the positioning enhancement data from the CLAS. The fixed solution calculation unit 135, using the PPP-RTK carrier phase positioning method, calculates a fixed solution based on the observation data and orbit data of the positioning satellite, the positioning enhancement data from the CLAS, and the integer offset.

[0110] <Action> Next, the operation of the positioning device according to Embodiment 2 will be described. In the operation of the positioning device according to Embodiment 2, Figure 2 The processing of the floating-point solution calculation unit 133 in step S206 and the processing of the fixed solution calculation unit 135 in step S210 are different from those in implementation method 1.

[0111] The floating-point solution calculation unit 133 and the fixed-point solution calculation unit 135 calculate the basic observation equations using the PPP-RTK method that uses the observation data of the rover station to replace the observation data of the base station, and using the positioning augmentation data and error model of CLAS. The observation equations yield the receiver-to-receiver single difference in phase and pseudorange of the master satellite and slave satellite, as shown in equations (14) and (15).

[0112] [Mathematical Expression 14]

[0113] [Mathematical Expression 15] ρ 1 B =r 1 B +Cδt B -CΔt 1 +I 1 B +T 1 B +ε ρB -{-ρ 1 A +r 1 A +C(δt A -Δt 1 )+I 1 A +T 1 A +ε ρA}…(15)

[0114] Equations (14) and (15) represent the inter-receiver phase difference and inter-receiver pseudo-range difference for observation data without using a reference station. Detailed explanations are omitted, but the calculation formulas for the inter-receiver differences used in practice are designed with reference to the CLAS user interface and example code (CLASLIB) released for the popularization of CLAS. Furthermore, the inter-satellite double difference regarding the phase and pseudo-range of the master and slave satellites is the difference between the inter-receiver differences of the master and slave satellites, which is the same as the VRS and RTK methods described in Implementation 1.

[0115] <Summary of Implementation Method 2> According to Embodiment 2 described above, the same effects as Embodiment 1 can be obtained. Furthermore, in Embodiment 2, by receiving positioning enhancement signals such as the L6 signal from a quasi-zenith satellite (a positioning enhancement satellite), both floating-point and fixed-point solutions can be obtained, thus suppressing operating costs such as communication fees. Moreover, quasi-zenith satellites have a wide domestic coverage area, and the access point does not need to be changed during wide-area operation; therefore, the positioning device can be designed without considering the access point.

[0116] <Variation Example> In implementation 2, the positioning enhancement signal receiver 12 receives positioning enhancement signals from positioning enhancement satellites, but is not limited to this. For example, such as Figure 9 As shown, the receiver can connect to the server (not shown) of the location enhancement signal distribution provider via Internet line 12a. Furthermore, authentication of the specified access point is performed during connection. Then, the location enhancement signal receiver 12 can receive location enhancement signals, including PPP-RTK location enhancement data, from the connected server at predetermined intervals. Even with this configuration, the same effects as in embodiment 2 can be achieved.

[0117] <Implementation Method 3> Figure 10 This is a block diagram showing the structure of the positioning device according to Embodiment 3. Hereinafter, structural elements that are the same as or similar to the above-described structural elements will be labeled with the same or similar reference numerals, and the different structural elements will be mainly described.

[0118] The structure of this embodiment 3 is the same as the structure obtained by adding a composite positioning part 14 to embodiment 2.

[0119] The structure of the satellite positioning unit 13 is roughly the same as that in Embodiment 2. However, in this Embodiment 3, the satellite positioning error prediction unit 136 predicts the positioning error of a single positioning solution based on the positioning error of the single solution and the difference between the single solution and the single positioning solution when either a floating-point solution or a fixed solution is obtained.

[0120] The composite positioning unit 14 includes a speed sensor 141, a distance measurement unit 142, a speed sensor correction unit 143, an angular velocity sensor 144, a yaw angle measurement unit 145, an angular velocity sensor correction unit 146, an autonomous navigation unit 147, a composite positioning unit 148, and a composite positioning error prediction unit 149.

[0121] Speed ​​sensor 141 outputs a pulse signal corresponding to the distance traveled by the vehicle. Distance measurement unit 142 calculates the travel distance and speed based on the number of pulses measured by speed sensor 141 in each specified cycle. Speed ​​sensor calibration unit 143 calculates the SF coefficient (proportioning factor) representing the distance of each pulse output by speed sensor 141.

[0122] The angular velocity sensor 144 uses the vertical direction of the vehicle as its detection axis and outputs a signal corresponding to the angular velocity (e.g., yaw rate) at the 0-point output. The yaw angle calculation unit 145 calculates the yaw angle based on the output of the angular velocity sensor 144 measured at each specified time interval. The angular velocity sensor calibration unit 146 calculates the 0-point output of the angular velocity sensor 144.

[0123] The autonomous navigation unit 147 updates its own position (hereinafter referred to as "DR position"), speed, and bearing (hereinafter referred to as "DR bearing") using the distance traveled by the distance measurement unit 142 and the yaw angle measured by the yaw angle measurement unit 145, according to autonomous navigation (Dead Reckoning). That is, the autonomous navigation unit 147 uses sensors such as the speed sensor 141 and the angular velocity sensor 144 to infer an autonomous navigation solution including the DR position. The autonomous navigation solution may include the vehicle speed and the DR bearing.

[0124] The composite positioning unit 148 obtains the individual positioning solution calculated by the individual positioning solution calculation unit 132 via the satellite positioning error prediction unit 136. It uses the positioning error (offset, validity period of a predetermined time) of the individual positioning solution calculated when a fixed or floating-point solution was obtained before a predetermined epoch to correct the positioning error of the individual positioning solution in the current epoch. Furthermore, the composite positioning unit 148 calculates the error of the autonomous navigation solution based on the autonomous navigation solution inferred by the autonomous navigation unit 147 and the individually positioning solution with corrected positioning errors. Based on this error, it corrects the autonomous navigation solution and performs composite positioning to obtain the composite positioning solution. The composite positioning solution includes not only the vehicle's position after error correction based on the autonomous navigation solution, but also the vehicle's speed and bearing after error correction based on the autonomous navigation solution.

[0125] The composite positioning error prediction unit 149 predicts the errors in the vehicle's position and orientation.

[0126] Furthermore, if either a floating-point solution or a fixed solution is obtained, the composite positioning unit 148 calculates the vehicle position based on the composite positioning solution, the positioning error of the composite positioning solution obtained by the composite positioning error prediction unit 149, and either the floating-point solution or the fixed solution used in predicting the positioning error of the composite positioning solution. Then, the composite positioning error prediction unit 149 predicts the positioning error of the composite positioning solution based on the positioning error of either the floating-point solution or the fixed solution, and the difference between either solution and the composite positioning solution.

[0127] <Action> Next, the operation of the positioning device according to Embodiment 3 will be explained. Figure 11 and Figure 12 This is a flowchart illustrating the actions of the positioning device for each epoch.

[0128] In step S1001, the processing of the positioning device is initialized.

[0129] Figure 11 The processing in steps S1002 to S1005 is performed by the composite positioning unit 14.

[0130] In step S1002, the distance measurement unit 142 multiplies the number of pulses of the speed sensor 141 measured in each specified cycle by the SF coefficient to calculate the moving distance, and uses the value after passing the number of pulses in each specified cycle through a low-pass filter to calculate the speed.

[0131] In step S1003, the angular velocity sensor correction unit 146 determines whether the vehicle is stopped based on the moving distance of the distance measurement unit 142, calculates the average value of the output of the angular velocity sensor 144 during the vehicle's parking, and uses this average value as the output bias of the angular velocity sensor 144 for correction. Furthermore, the processing in step S1003 can, for example, utilize the processing described in Japanese Patent No. 3137784 or Japanese Patent No. 3751513.

[0132] In step S1004, the yaw angle measurement unit 145 calculates the yaw angle obtained by removing the output bias from the output of each angular velocity sensor 144 measured at a specified time.

[0133] In step S1005, the autonomous navigation unit 147 calculates the movement vector for each specified period based on the movement distance and yaw angle according to the autonomous navigation, and updates the vehicle position by adding the calculated movement vector to the vehicle position measured last time.

[0134] The processing in steps S1006 to S1015 is performed by the satellite positioning unit 13. These processes are related to... Figure 2 The processes from steps S202 to S211 are the same, so the explanation is omitted.

[0135] Figure 12 In the process, step S1019 is performed by the satellite positioning unit 13, while steps S1016 to S1026, excluding step S1019, are performed by the composite positioning unit 14.

[0136] In step S1016, the composite positioning unit 14 determines whether a non-positioning solution is set to a positioning solution state. If a non-positioning solution is not set, the process proceeds to step S1017; if a non-positioning solution is set, the process proceeds to step S1022.

[0137] In step S1017, the composite positioning unit 148 performs composite positioning based on the autonomous navigation solution, the individual positioning solution, and the pseudo-range, thereby obtaining a composite positioning solution that can match the driving trajectory with an accuracy of approximately 2 meters even in the presence of local radio wave obstruction and multipath. Furthermore, the calculation of the composite positioning solution can, for example, utilize the composite positioning methods described in Japanese Patent No. 6482720, Japanese Patent No. 4988028, and Japanese Patent No. 5855249.

[0138] In step S1018, the composite positioning unit 14 determines whether either the floating-point solution or the fixed solution is set to the positioning solution state. If either the floating-point solution or the fixed solution is set, the process proceeds to step S1019. If neither the floating-point solution nor the fixed solution is set, that is, if only the positioning solution is set, the process proceeds to step S1022.

[0139] In step S1019, the satellite positioning error prediction unit 136 predicts the positioning error (also known as offset) of a single positioning solution based on the positioning error of the solution obtained from the floating-point solution and the fixed solution, and the difference between the obtained solution and the single positioning solution.

[0140] Figure 13 This is a diagram used to illustrate the prediction of the positioning error of a single positioning solution. The following explanation assumes the solution is a fixed solution, but the same applies even if the solution is a floating-point solution. Figure 13 (a) shows a sphere containing a fixed solution. Figure 13 (b) shows a local horizontal plane that is part of the sphere, a point P1 representing a single localized solution, and a point P2 representing a fixed solution obtained in the same epoch as the single localized solution.

[0141] The satellite positioning error prediction unit 136 predicts the error of a single positioning solution based on the predicted positioning error (i.e., prediction error) of the fixed solution and the difference between point P1 representing the single positioning solution and point P2 representing the fixed solution (i.e., the distance between the two points, δx, δy, δz). For example, the satellite positioning error prediction unit 136 uses the prediction error value, where a smaller distance between the two points indicates a closer approximation to the fixed solution, as the error of the single positioning solution. Furthermore, the error of the single positioning solution predicted here is used to fine-tune the composite positioning solution based on the previous epoch during the next composite positioning.

[0142] Figure 12 In step S1020, the composite positioning error prediction unit 149 predicts the positioning error of the composite positioning solution based on the positioning errors of the solutions obtained from the floating-point solution and the fixed solution, and the difference between the obtained solutions and the composite positioning solution. The prediction of the positioning error of the composite positioning solution is performed in the same way as the prediction of the positioning error of the individual positioning solution described in step S1019. For example, the composite positioning error prediction unit 149 predicts the error of the composite positioning solution based on the prediction error of the fixed solution and the difference between the point representing the composite positioning solution and the point representing the fixed solution, i.e., the distance between the two points. Furthermore, the error of the composite positioning solution predicted here is used not only for updating the composite positioning error in the next epoch, but also for weighting in the gain calculation of the extended Kalman filter for the individual positioning solution or pseudo-range.

[0143] In step S1021, the composite positioning unit 148 selects the solution with higher accuracy and matching to the driving trajectory from the floating-point solution, fixed solution, and autonomous navigation solution. For example, the composite positioning unit 148 selects the solution with the smallest error among the floating-point solution, fixed solution, and autonomous navigation solution. Then, the process proceeds to step S1024.

[0144] In step S1022, the composite positioning unit 14 determines whether the error of the composite positioning solution was predicted (calculated) in step S1020 within the most recent specified time or specified distance. If the error of the composite positioning solution was predicted, the process proceeds to step S1023; otherwise, the process proceeds to step S1024.

[0145] In step S1023, the composite positioning unit 148 calculates the position of the vehicle based on the composite positioning solution, the positioning error of the composite positioning solution obtained by the composite positioning error prediction unit 149, and any one of the floating-point solution and fixed solution used in the calculation of the positioning error of the composite positioning solution.

[0146] Figure 13 (c) is a diagram illustrating the process of step S1023. Figure 13 In (c), point P3 represents the past composite positioning solution synchronized with satellite positioning used in the error calculation of the composite positioning solution, point P4 represents the composite positioning solution at the latest time, and point P5 represents the fixed solution used in the error calculation of the composite positioning solution. The composite positioning unit 148 corrects point P3 of the composite positioning solution to point P5 of the fixed solution used in the error calculation of the composite positioning solution, and corrects the composite positioning solution at the latest time from point P4 to point P6 based on this correction, thereby determining point P6 as the position of the vehicle. After that, the process proceeds to step S1024.

[0147] In step S1024, the composite positioning error prediction unit 149 continues to predict the vehicle's position error within the most recent specified time or distance, regardless of which of the floating-point solution, fixed solution, autonomous navigation solution, and composite positioning solution was used to update the vehicle's position. Furthermore, the vehicle's position error is predicted based on a high-precision positioning solution with a small prediction error.

[0148] In step S1025, the speed sensor calibration unit 143 calibrates the SF coefficient of the pulse signal of the speed sensor. This calibration can, for example, use the calibration method described in Japanese Patent No. 5606656.

[0149] In step S1026, the angular velocity sensor correction unit 146 uses the vehicle's orientation at any moment during the vehicle's movement as its initial value. Based on the difference between the orientation obtained by accumulating the yaw angle at each moment and the vehicle's orientation corrected by the composite positioning unit 148, it corrects the 0 point (also called offset) of the angular velocity sensor 144. Furthermore, this correction can, for example, use the correction methods described in Japanese Patent No. 3321096 and Japanese Patent No. 3727489. After step S1026, the process returns. Figure 11 Step S1001.

[0150] <Summary of Implementation Method 3> Generally, unless radio waves are completely cut off like in a tunnel, the individual positioning solution from dual-frequency multi-GNSS satellites can be obtained with a high positioning rate. Therefore, an appropriate individual positioning solution using pseudo-range with less multipath influence can be obtained. The accuracy of the individual positioning solution is only a few meters, so it is not high precision. However, if it is calculated using a three-dimensional velocity vector based on Doppler, it becomes a trajectory close to autonomous navigation, as long as it is not a large multipath environment such as buildings and streets.

[0151] Here, according to the positioning device of Embodiment 3, when obtaining either a floating-point solution or a fixed solution, the positioning error (offset) of the individual positioning solution whose trajectory shape matches the vehicle's driving trajectory is obtained based on the positioning error of the individual solution and the difference between the individual solution and the individual positioning solution. As a result, the individual positioning solution after the positioning error is corrected gradually approaches the floating-point solution or the fixed solution, thus improving the positioning rate, accuracy, and usability of the individual positioning solution.

[0152] Furthermore, according to Embodiment 3, even if no floating-point or fixed solution is obtained in the next epoch, the positioning error obtained in the previous epochs can be used to correct the individual positioning solution in the next epoch. Therefore, the high accuracy of the individual positioning solution, which gradually approaches the floating-point or fixed solution, is maintained, and the positioning rate, accuracy, and usability of the corrected individual positioning solution are improved. Moreover, the accuracy of the composite positioning solution corrected using the individual positioning solution with the corrected positioning error is also improved.

[0153] Furthermore, according to Embodiment 3, when either a floating-point solution or a fixed solution is obtained, the positioning error of the composite positioning solution is predicted based on the positioning error of either solution and the difference between either solution and the composite positioning solution. The composite positioning solution, after correcting the positioning error, gradually approaches the floating-point solution or the fixed solution; therefore, the positioning rate, accuracy, and usability of the composite positioning solution are further improved.

[0154] <Variation Example 1> The Doppler prediction value based on autonomous navigation can be returned to the satellite positioning unit 13. If it is during the radio wave cutoff process within a specified time, the prediction of pseudo-range can continue. Therefore, the reliability of positioning using low-quality pseudo-range immediately after radio wave cutoff is improved, and the occurrence of erroneous positioning can be further suppressed.

[0155] <Variation Example 2> Figure 14 , Figure 15 and Figure 16 This diagram illustrates the situation where the prediction errors of floating-point and fixed solutions are further determined by comparing the trajectories of floating-point and fixed solutions with the trajectory of autonomous navigation. Figure 15 It is magnification Figure 14 The image after AR1 in the region, Figure 16 It is magnification Figure 14 The image after AR2 of the region. Figures 14-16 The diagram appropriately illustrates the vehicle's position 1, fixed solution 2, prediction error 3, driving path 5, trajectory 6 based on autonomous navigation (DR position), and trajectory 7 after performing an affine transformation on trajectory 6.

[0156] The positioning device can detect situations where the actual error is sufficiently large relative to the predicted error by comparing the trajectories of floating-point and fixed solutions with the autonomous navigation trajectory. This structure can cover situations where the accuracy of the floating-point solution's prediction error decreases, resulting in a safer usage method. Furthermore, when there is a mismatch with the driving trajectory, by significantly re-evaluating the prediction errors of both floating-point and fixed solutions, the device can more actively use the floating-point and fixed solutions that match the driving trajectory.

[0157] <Variation Example 3> The above describes the structure associated with the nearest fixed and floating-point solutions with fewer errors, and the composite positioning solution after correcting autonomous navigation or prediction errors, but it is not limited to this. For example, by maintaining autonomous navigation to achieve fixed and floating-point solutions that pass through multiple nearest locations with fewer errors, the orientation of autonomous navigation becomes more accurate, high precision can be maintained for a long time, and its usability in vehicular environments is improved.

[0158] <Implementation Method 4> Figure 17 This is a block diagram showing the structure of the driving assistance system according to Embodiment 4. Figure 17 The driving assistance system includes the positioning device according to Embodiment 3. As described below, based on the high-precision positioning results obtained by the positioning device, it displays the vehicle's position within the lane and provides lane guidance to the driver and passengers.

[0159] Figure 17In addition to the positioning device, the driving assistance system also includes high-precision map data 15, mapping and matching unit 16, information output unit 17, display unit 18, operation input unit 19 and driving assistance control unit 41.

[0160] The high-precision map data 15 is generated with an absolute precision of less than 50 cm, and includes three-dimensional shape information of each lane, three-dimensional shape information of the road shoulders, their longitudinal and lateral slopes, and road elevation information. The mapping and matching unit 16 performs lane unit mapping and matching based on the vehicle position, vehicle orientation, and their prediction errors obtained by the composite positioning unit 14, and the high-precision map data 15, and determines the lane in which the vehicle 1 is traveling, i.e., the driving lane, and the vehicle position within the lane. In addition, the mapping and matching can, for example, use the mapping and matching described in Japanese Patent No. 6482720.

[0161] The information output unit 17 generates ADAS (Advanced Driving Assistance System) data according to the Advanced Driving Assistance Systems Interface Specifications (ADASIS) standard and outputs it to the display unit 18 and the driver assistance control unit 41. Furthermore, the information output unit 17 generates ADAS data based on the vehicle's position, orientation, and their errors obtained by the composite positioning unit 14, the driving lane and the vehicle's position within the lane determined by the mapping matching unit 16, and the corresponding data of the specified distance along the road ahead of the vehicle 1 from the high-precision map data 15.

[0162] The operation input unit 19, for example, is an input button, which accepts input operations to reflect the driver's and passengers' intentions on the display screen. The display unit 18 generates an image of the display screen based on ADAS data from the information output unit 17 and high-precision road data around the vehicle 1 from the high-precision map data 15. It overlays the vehicle's position within the lane and other parameters on this image, and displays it, or provides voice guidance. Additionally, based on the input operations received from the driver and passengers by the operation input unit 19, the display unit 18 displays the selected display screen, or changes the scale and content of the display screen. The driver assistance control unit 41 generates lane guidance data for the destination based on the input operations from the driver and passengers and the ADAS data from the information output unit 17, and displays this data on the display unit 18 or provides voice guidance.

[0163] Next, use Figures 18-20 Here is an example of the display screen of the display unit 18. Figures 18-20 This is an example diagram showing a display indicating the vehicle's position within the lane. Figure 18 In this diagram, the lateral distances between the vehicle and the white lines at both ends of the driving lane are shown, defining the vehicle's position. Distance d1 is the width of the driving lane, i.e., the distance between the white lines. Distance d2 is the distance between the center of the driving lane and the center of the vehicle. Distance d3 is the distance between the left white line and the vehicle, and distance d4 is the distance between the right white line and the vehicle. Distances d3 and d4 are calculated using a predetermined lateral width of the vehicle.

[0164] Figure 19 In this process, the display unit 18 changes the display color of the strip portions 51a and 51b at distances d3 and d4 based on whether they are the same. For example, if distances d3 and d4 are the same, the strip portions 51a and 51b are displayed with the same color; if distances d3 and d4 are different, then... Figure 19 As shown, the strip portion 51a and the strip portion 51b are displayed in different colors.

[0165] Figure 20 In this configuration, the display unit 18 changes the display color of the bar 51c and the arrow 51d based on at least one of distances d2, d3, and d4. For example, when the distance between the vehicle 1 and the white line is greater than or equal to a first threshold, the bar 51c and the arrow 51d are displayed in green. Similarly, when the distance is less than the first threshold but greater than a second threshold, the bar 51c and the arrow 51d are displayed in yellow, and when the distance is less than the second threshold, the bar 51c and the arrow 51d are displayed in red.

[0166] <Summary of Implementation Method 4> According to the driving assistance system described in Embodiment 4 above, the accurate driving lane and the vehicle's position within the lane can be determined using high-precision positioning results obtained from satellite positioning and composite positioning. Therefore, the gap between the left and right white lines of the driving lane and the vehicle 1 can be represented by an easily understandable image, thereby showing the driver and passengers the most recent driving conditions. Thus, for example, when the possibility of dangerous driving due to drowsy driving or lack of attention to the road increases, attention can be accurately drawn and warnings issued, thus easily preventing situations where the vehicle 1 deviates from the lane and road, making it difficult to continue driving, and preventing self-damage accidents.

[0167] <Variation Example> use Figures 21-28 Here is an example illustrating the distortion of the display screen on the display unit 18. As will be clearly stated in the following description, for any example, by using the result of high-precision positioning, effects such as driving assistance and safety prevention can be obtained.

[0168] Figures 21-23This is an example diagram showing a display screen showing the driving trajectory corresponding to the most recent specified time (or specified distance). In this vehicle 1, as shown... Figure 21 When driving in a winding manner as shown, or when the vehicle 1 is as shown Figure 22 As the vehicle gradually approaches the white line as shown, the display unit 18 displays the vehicle's trajectory on the lane, or draws attention through visual display and sound. Furthermore, the display unit 18 is based on... Figure 23 The white lines (dashed and solid lines) used by vehicle 1 to change lanes are shown to indicate the vehicle's trajectory in the lane or to change the content that draws attention.

[0169] Based on this structure, the high-precision positioning results obtained from satellite positioning and composite positioning are used to show the driving trajectory of vehicle 1 on the lane, thereby showing the driver and passengers the most recent driving conditions. Thus, for example, when the possibility of dangerous driving due to drowsy driving or lack of attention to the road ahead increases, attention can be accurately drawn and warnings issued. Therefore, situations where vehicle 1 deviates from the lane and road and becomes difficult to continue driving, as well as situations where self-damage accidents occur, can be easily avoided.

[0170] Figure 24 and Figure 25 This diagram illustrates an example of a display screen used as a display unit 18 when a head-up display is installed near the windshield of the vehicle 1, which is easily accessible to the driver. (See diagram for example.) Figure 24 Thus, in situations where visibility is poor due to darkness at night, heavy rain, dense fog, or blizzards, and such as Figure 25 When the white line is not visible due to snow accumulation, the display unit 18 displays the positional relationship between the boundary line 52a and the white line 52b of the driving lane and the vehicle 1.

[0171] Based on this structure, even in situations of poor visibility or when white lines cannot be recognized, the high-precision positioning results obtained from satellite positioning and composite positioning can be used to provide AR (Augmented Reality) assistance, which accurately informs the driver and passengers of the driving lane and the vehicle's position within the lane. Therefore, even in conditions of poor visibility or when white lines cannot be recognized, proper driving can be maintained, easily avoiding situations where the vehicle deviates from its lane and road, making continued driving difficult, or causing self-damage accidents.

[0172] Figure 26 This is an example diagram showing a display indicating the nearest emergency stopping lane 53 on the highway ahead of the vehicle 1. The display 18 guides the driver and passengers to the location of the nearest emergency stopping lane 53 and provides lane guidance towards it before the driver and passengers become unwell and unable to continue driving safely for an extended period.

[0173] Based on this structure, high-precision positioning results obtained from satellite positioning and composite positioning can be used to guide the vehicle to the nearest emergency stopping lane 53. This improves the reliability of lane guidance to the emergency stopping lane 53.

[0174] Figure 27 This diagram shows an example of a display screen indicating the destination lane ahead of the vehicle 1. The display unit 18 displays an image 54 of the destination in each lane of the road in which the vehicle 1 is traveling. In addition, when the lane to the destination is determined, the display unit 18 may only display the image 54 of that lane, or it may display a prompt to change lanes to the destination and provide voice guidance.

[0175] Based on this structure, high-precision positioning results obtained from satellite positioning and composite positioning can be used for lane guidance to the destination. This improves the reliability of lane guidance to the destination.

[0176] Figure 28 This is an example of a display screen showing a scene when the vehicle 1 enters a lane of a general road or other road in which the vehicle 1 cannot pass. When the vehicle 1 is about to enter (go against traffic) a lane in which the direction of travel is prohibited according to traffic rules, or when it has already entered (go against traffic), the display unit 18 displays an image 54 and other displays to attract attention and provide warnings and voice guidance.

[0177] Based on this structure, even in the center of an intersection where white lines are not displayed, the high-precision positioning results obtained from satellite positioning and composite positioning can be used to check (detect) the state of the vehicle 1 starting to drive in the wrong direction. Therefore, dangerous driving caused by the driver's inattention can be predicted as early as possible. Thus, attention can be drawn and warnings issued at precise timing, rather than drawing attention and issuing warnings to the driver and passengers unprepared.

[0178] <Implementation Method 5> Figure 29 This is a block diagram illustrating the structure of the driving assistance system according to Embodiment 5. Figure 29 The driving assistance system and the implementation method 4 Figure 17 The structure is the same as that obtained by adding a vehicle perimeter measurement unit 21, a front stereo camera 22, a front millimeter-wave radar 23, a rear stereo camera 24, a left front millimeter-wave radar 25, a right front millimeter-wave radar 26, a left rear millimeter-wave radar 27, and a right rear millimeter-wave radar 28. Figure 29 The driver assistance system, as described below, can detect the presence and movement of obstacles such as other vehicles around the vehicle.

[0179] Figure 30 and Figure 31These are side and top views showing the installation positions and measurement ranges of the stereo cameras 22 and 24 and the millimeter-wave radars 23, 25, 26, 27, and 28.

[0180] Stereo cameras 22 and 24 are respectively positioned on the upper parts of the front and rear windshields, and have detection ranges 22a and 24a respectively. Detection ranges 22a and 24a have detection angles of 40 degrees relative to the front and rear, respectively, and detection distances of 100m and 40m, respectively. Furthermore, the detection angle and detection distance values ​​described here are examples and are not limited to these.

[0181] Millimeter-wave radar 23 is positioned in the center of the front bumper and has a detection range 23a. The detection range 23a has a narrow detection angle of 20 degrees relative to the front and a relatively long detection distance of 200m. However, the detection angle and detection distance values ​​described herein are examples and are not limited to these. Millimeter-wave radars 25 and 26 are positioned in the corners of the front bumper and have detection ranges 25a and 26a, respectively. Millimeter-wave radars 27 and 28 are positioned in the corners of the rear bumper and have detection ranges 27a and 28a, respectively. The detection ranges 25a, 26a, 27a, and 28a have a wide detection angle of 120 degrees relative to the left front, right front, left rear, and right rear, respectively, and a relatively short detection distance of 30m. Furthermore, the detection angle and detection distance values ​​described herein are examples and are not limited to these.

[0182] Next, the characteristics of a stereo camera will be explained. A stereo camera consists of left and right cameras that capture images of various obstacles such as other vehicles and pedestrians, as well as road markings such as white and yellow lines. Based on the deviation (parallax) of these images, it detects the three-dimensional position, size, and shape of the boundaries (brightness variations) of obstacles and road markings. Even if an obstacle moves across these detection directions, the stereo camera can still detect its movement. However, if there is dirt or fog on the windshield in front of the stereo camera lens, or if the vehicle's lights are not on in inclement weather (heavy rain), backlighting, nighttime, or in tunnels, the detection performance of the stereo camera will decrease.

[0183] Next, the characteristics of millimeter-wave radar will be explained. Millimeter-wave (electromagnetic) waves transmitted within a predetermined detection angle are reflected off an obstacle and back to the millimeter-wave radar, which then detects the distance between the radar and the obstacle. Millimeter-wave radar exhibits excellent long-range ranging performance and can maintain its ranging performance regardless of sunlight conditions, brightness, or weather (rain, fog). However, millimeter-wave radar has difficulty detecting obstacles with low reflectivity and the movement of obstacles crossing the radar's detection direction.

[0184] In view of the above, in this implementation method 5, Figure 29The vehicle perimeter measurement unit 21 collaboratively identifies and integrates the presence and movement of obstacles detected by stereo cameras 22 and 24 and millimeter-wave radars 23, 25, 26, 27, and 28. In other words, the vehicle perimeter measurement unit 21 monitors the area around the vehicle 1 by combining the detection results from the stereo cameras and millimeter-wave radars.

[0185] Next, an example of surveillance around the vehicle 1 performed by stereo cameras 22 and 24 and millimeter-wave radars 23, 25, 26, 27, and 28 will be described. Figures 32-34 This diagram illustrates how the front stereo camera 22 and the front millimeter-wave radar 23 detect the movements of other vehicles 56a and 56b in front of this vehicle 1.

[0186] like Figure 32 As shown, the stereo camera 22 has a wide detection angle but a short detection distance. Therefore, Figure 32 In the example, the stereo camera 22 detects other vehicles 56a traveling in the left adjacent lane of the vehicle 1's driving lane within the detection distance, but does not detect other vehicles 56b traveling outside the detection distance of the vehicle 1's driving lane.

[0187] In addition, the stereo camera 22 can detect the type of white line as a dashed line by detecting the shape of the left and right white lines (the shaded parts in the figure) of the driving lane within the detection range 23a.

[0188] In addition, such as Figure 33 As shown, the stereo camera 22 can detect the shape of obstacles such as other vehicles 56a and the shape of the white line (the left side of the driving lane), and therefore can also detect the distance d6 between the obstacle and the white line.

[0189] In contrast, such as Figure 34 As shown, the millimeter-wave radar 23 has a narrow detection angle but a long detection range. Therefore, Figure 34 In the example, the millimeter-wave radar 23 does not detect other vehicles 56a traveling in the left adjacent lane of the driving lane of vehicle 1, but detects other vehicles 56b traveling in the driving lane of vehicle 1.

[0190] Vehicle Peripheral Measurement Unit 21 Based on Figures 32-34 The detection results are used to detect the distance and relative position between the vehicle 1 and other vehicles 56b traveling in the driving lane and other vehicles 56a traveling to the left of the driving lane, and to detect the type of white line in the driving lane.

[0191] <Summary of Implementation Method 5> According to the driving assistance system described in Embodiment 5 above, the high-precision positioning result (vehicle position) obtained from satellite positioning and composite positioning, along with obstacles in front of the vehicle 1 detected by a stereo camera and millimeter-wave radar, can be mapped onto a high-precision map of lane units. Therefore, the traffic conditions in front of the vehicle 1 can be included, displaying the driving status of the vehicle 1 to the driver and passengers. Thus, for example, when the possibility of dangerous driving due to drowsy driving or lack of attention to the road increases, attention can be accurately drawn and warnings issued.

[0192] <Variation Example> Figure 35 This is an example diagram used to illustrate how the type of white line in the driving lane is determined based on the stereo camera 22.

[0193] The stereo camera 22 can detect whether the white lines on the left and right sides of the driving lane are solid or dashed. Therefore, the vehicle perimeter measurement unit 21 can infer whether the driving lane is the left lane, the right lane, or an inner lane other than those detected by the stereo camera 22 based on the combination of the types of white lines on the left and right sides.

[0194] Furthermore, when the white line is dashed, the vehicle perimeter measurement unit 21 can determine the boundary (outline) of a continuous driving lane by linearly interpolating the blank areas between the preceding and following white sections using the preceding and following white sections. In Japan, in dashed white lines on highways, an 8m white section and a 12m blank section overlap, while in dashed white lines on ordinary roads, a 5m white section and a 5m blank section overlap. Therefore, the vehicle perimeter measurement unit 21 can detect whether the vehicle 1 is traveling on a highway or an ordinary road based on the interval between the white and blank sections.

[0195] The driver assistance control unit 41 compares the high-precision positioning results obtained by satellite positioning and composite positioning with the road information detected by the vehicle perimeter measurement unit 21 to determine the lane with higher reliability as the driving lane that the vehicle 1 should travel in, and guides the vehicle to change lanes to the destination.

[0196] According to this structure, not only can the vehicle's position and driving lane information obtained by the satellite positioning unit 13, the composite positioning unit 14, and the mapping and matching unit 16 be used, but also the driving lane information detected by the vehicle perimeter measurement unit 21 can be used. Therefore, by utilizing the advantages of each piece of information, the performance (e.g., accuracy and reliability) of the driver assistance control unit 41 regarding the driving lane can be improved, thus providing stable driving assistance to the driver and passengers.

[0197] Figure 36This is an example diagram used to illustrate the detection of protrusions 57a and concave parts 57b on the road surface, as well as obstacles 57c on the road. The front stereo camera 22 detects protrusions 57a such as concrete embedment, concave parts 57b such as pavement damage, and obstacles 57c such as lost objects within the detection range 22a.

[0198] The driver assistance control unit 41 determines whether ride comfort is affected based on the detection results of the convex part 57a, concave part 57b and obstacle 57c by the stereo camera 22. Then, based on the determination result, the driver assistance control unit 41 controls the display and sound of the display unit 18 to convey to the driver and passengers that ride comfort has been affected, as well as to draw their attention to avoid obstacles and issue warnings.

[0199] According to this structure, the driver and passengers can know in advance the protrusions 57a and concave parts 57b of the road surface and obstacles 57c on the road, so they can drive appropriately near the location, such as slowing down or driving carefully.

[0200] Figures 37-40 This is a diagram illustrating an example of surveillance around vehicle 1 when another vehicle 56c overtakes vehicle 1 and changes lanes in front of vehicle 1, arranged chronologically. Figure 37 In the middle, the millimeter-wave radar 28 on the right rear detects other vehicles 56c traveling on the right rear of this vehicle 1. Figure 38 In the middle, the millimeter-wave radar 26 on the right front detects other vehicles 56c that are driving on the right front and have overtaken this vehicle 1. Figure 39 In the middle, the front stereo camera 22 detects other vehicles 56c that are changing lanes from the right front of vehicle 1 to the front of vehicle 1. Figure 40 In the middle, the front stereo camera 22 and the front millimeter-wave radar 23 detect other vehicles 56c traveling in front of this vehicle 1.

[0201] During this operation, the driver assistance control unit 41 maps obstacles around the vehicle 1 detected by multiple stereo cameras and multiple millimeter-wave radars, centered on the vehicle's position on a high-precision map (e.g., on a road where the vehicle 1 is traveling). Then, the driver assistance control unit 41 predicts the movement of each obstacle at the next measurement time, and when the next measurement time arrives, determines whether the obstacles are moving roughly as predicted, whether a new obstacle is detected, or whether previously detected obstacles are gradually moving away. For example, as... Figures 37-40 In this case, when multiple stereo cameras and multiple millimeter-wave radars detect other vehicles 56c, the driver assistance control unit 41 determines that the other vehicles 56c are moving roughly as predicted.

[0202] In general, besides vehicle 1, there are various obstacles on the road, including other vehicles, two-wheeled vehicles, bicycles, and pedestrians, some of which are stationary and others that move independently. With the driving assistance system configured as described above, the driver and passengers are aware of the presence of obstacles around vehicle 1, the positional relationship between the obstacles and vehicle 1, and the movement of the obstacles, thus easily avoiding collisions with them.

[0203] Furthermore, as shown in (1) to (3) below, even if a part of the structure is different from the structure of Embodiment 5, the same effect as the structure of Embodiment 5 can be obtained.

[0204] (1) Either a stereo camera or millimeter-wave radar can be replaced with an ultrasonic sensor. An ultrasonic sensor reflects emitted ultrasonic waves back to the object being detected, and can detect the presence of the object and the distance to obstacles by receiving the reflected waves. The detection sensitivity of an ultrasonic sensor does not depend on the reflectivity of the object being detected, it is highly resistant to dust and dirt, and can detect transparent objects such as glass, as well as complex objects such as metal mesh, and it is inexpensive. However, sound waves travel slower than electromagnetic waves, and the detection range of an ultrasonic sensor is 100m, shorter than that of millimeter-wave radar. Therefore, ultrasonic radar can be used as a sonar for parking.

[0205] (2) Either a stereo camera or millimeter-wave radar can be replaced with LiDAR (Light Detection and Ranging). LiDAR reflects emitted laser light (infrared) towards the target object and, by receiving the reflected waves, can accurately detect not only the distance to the target object but also its position and shape. Because it uses infrared light, which has a shorter wavelength than electromagnetic waves, LiDAR can detect obstacles smaller than millimeter-wave radar and can detect obstacles with low reflectivity. As mentioned above, LiDAR is preferred for applications requiring more accurate detection of the shape and position of obstacles. However, the higher cost of LiDAR compared to millimeter-wave radar and the reduced detection capability in inclement weather should be noted.

[0206] (3) When the driver assistance system detects Figure 36In cases involving road surface protrusions 57a and concave areas 57b, as well as road obstacles 57c, the detection results can be sent to a road maintenance center for road surface maintenance. For example, when a road surface protrusion 57a ​​is detected, the driver assistance control unit 41 can automatically communicate with a pre-determined road maintenance center's server regarding the coordinates of the detected area and the detection results. Then, the road maintenance center can issue maintenance instructions to maintenance vehicles by organizing the coordinates and types of road surface protrusions 57a and the like recorded on the server, and generating a maintenance priority order.

[0207] <Implementation Method 6> Figure 41 This is a block diagram showing the structure of the driving assistance system according to Embodiment 6. Figure 41 The driving assistance system and the implementation method 5 Figure 29 The structure is the same after adding the road condition measurement unit 31, the road condition sensor 32, and the radar vehicle height gauge 33.

[0208] Figure 42 This is a side view showing the mounting positions of the road condition sensor 32 and the laser altitude meter 33. Additionally, Figure 42 The measurement directions 32a and 33a of the road condition sensor 32 and the laser vehicle height gauge 33 are shown in the figure, respectively.

[0209] The road surface condition sensor 32 detects (monitors) the road surface condition directly beneath the vehicle 1. For example, the road surface condition sensor 32 emits multi-wavelength near-infrared laser light onto the road surface and measures the reflection from the road surface to detect the road surface condition, including road surface roughness, and the thickness of various layers of dry, wet, frozen, and compacted snow. The laser height gauge 33 illuminates the road surface with a laser at an angle, receives the reflected light from the road surface, and uses triangulation to detect the distance (vehicle height) between the road surface and the vehicle 1, or to detect cracks, potholes, ruts, and unevenness of the road surface.

[0210] The road surface condition measurement unit 31 determines the overall road surface condition based on the detection results of the road surface condition sensor 32 and the laser vehicle height gauge 33.

[0211] The driver assistance control unit 41, based on the road condition determination result obtained by the road condition measurement unit 31, causes the display unit 18 to display and provide voice guidance to convey information such as ride comfort and driving impact to the driver and passengers. Furthermore, if the road condition determination result obtained by the road condition measurement unit 31 indicates that there is an obstacle to driving the vehicle 1, the driver assistance control unit 41 causes the display unit 18 to display and provide voice guidance to convey information such as alerts and warnings to the driver and passengers regarding avoidance.

[0212] <Summary of Implementation Method 6> According to the driving assistance system of this embodiment 6 as described above, the driver and passengers can know the road conditions in advance, and therefore can drive appropriately near the location, such as reducing speed or driving cautiously.

[0213] <Variation Example> The driving assistance system can send road conditions to a road maintenance center for road maintenance. For example, if it is determined that there is an obstacle to driving the vehicle 1, the driving assistance control unit 41 can automatically contact a pre-determined server of the road maintenance center with the coordinates of the determination and the determination result. Then, the road maintenance center can issue maintenance instructions to the maintenance vehicles by organizing the coordinates of the road conditions recorded on the server and generating a maintenance priority.

[0214] <Implementation Method 7> Figure 43 This is a block diagram showing the structure of the driving assistance system according to Embodiment 7. Figure 43 The driving assistance system and the implementation method 6 Figure 41 The structure is the same after adding vehicle control unit 42, drive control unit 43, brake control unit 44, and steering control unit 45.

[0215] The driver assistance control unit 41 controls the vehicle control unit 42 based on information from the information output unit 17, the operation input unit 19, the vehicle perimeter measurement unit 21, and the road condition measurement unit 31.

[0216] The vehicle control unit 42 has functions such as LKA (Lane Keeping Assist), LCA (Lane Change Assist), ACC (Adaptive Cruise Control), and AEB (Advanced Emergency Braking). Under the control of the driver assistance control unit 41, the vehicle control unit 42 uses the drive control unit 43, brake control unit 44, and steering control unit 45 to control the engine, brakes, and steering system.

[0217] The drive control unit 43 controls the drive system by adjusting the engine's fuel injection and selecting a gear corresponding to the vehicle's speed. For example, in situations where there is a risk of collision if the driver does not apply the brakes, the brake control unit 44 activates the brakes. The steering control unit 45 operates the steering mechanism to control the direction of travel of the vehicle 1.

[0218] <Summary of Implementation Method 7> According to the driving assistance system of Embodiment 7 described above, the high-precision positioning results obtained by satellite positioning and composite positioning can be used to appropriately control the driving of the vehicle 1. Next, several examples of the control of the vehicle 1 will be described.

[0219] <Example of vehicle control in Implementation Method 7> Figure 44 This diagram illustrates an example of driving control to prevent vehicle 1 from leaving the lane. Additionally, Figure 44 In the following figures, a double-dotted line is shown inside the white lines on the left and right sides of the driving lane at a predetermined distance d11.

[0220] When the driver requests lane keeping assist via the operation input unit 19, the driver assistance control unit 41 controls the drive control unit 43, brake control unit 44, and steering control unit 45 by controlling the vehicle control unit 42. Through this control, the vehicle 1 maintains a safe distance from other vehicles ahead and a predetermined distance d11 from the left and right white lines of the driving lane.

[0221] Additionally, the display unit 18 notifies the driver and passengers whether lane keeping assist is activated via display and sound. Furthermore, when lane keeping assist is activated, if the stereo camera 22 detects dashed white lines on either side of the driving lane, the vehicle perimeter measurement unit 21 calculates the boundaries (outlines) of the continuous driving lane by linearly interpolating the blank areas of the dashed lines using the white areas in front and behind. The information output unit 17 outputs the coordinates of the lane boundaries determined by the mapping matching unit 16, the type of driving lane (solid or dashed), and the vehicle's position to the driver assistance control unit 41. The driver assistance control unit 41 compares the coordinates of the driving lane boundaries, the type of driving lane (dashed or solid), and the vehicle's position with the driving lane boundary lines obtained by the vehicle perimeter measurement unit 21 to control the distance between the vehicle's position and the left and right white lines of the driving lane.

[0222] According to this structure, the driver assistance control unit 41 uses the coordinates representing the lane boundary determined by the mapping matching unit 16 and the vehicle's position from the information output unit 17 to assist the vehicle 1 in maintaining its lane. Therefore, for example, even if the distance between the vehicle 1 and other vehicles ahead decreases, and the front stereo camera 22 is temporarily unable to detect the left and right white lines of the driving lane, as long as the prediction error of the vehicle's position remains below a predetermined error, stable lane-keeping assistance can be provided to the driver and passengers.

[0223] Figure 45 and Figure 46This diagram illustrates an example of determining whether a lane change is permissible based on the type of white lines (dashed and solid) on the left and right sides of the driving lane in front of vehicle 1.

[0224] Figure 45 In this case, the white lines on the left and right sides of the driving lane of vehicle 1 are solid lines that prohibit vehicles from crossing. Even if the operation input unit 19 receives the operation to implement the lane change assist function from the driver, the driver assistance control unit 41 retains the command to change the driving lane and outputs the command to maintain the driving lane to the lane control unit 42.

[0225] Figure 46 In this configuration, the white line on the right side of the driving lane of vehicle 1 is a dashed line that the vehicle can cross. In this case, if the operation input unit 19 receives an operation from the driver to implement the lane change assist function, the driver assistance control unit 41, after confirming that there are no obstacles around vehicle 1, generates a driving path 58 including coordinates and orientation. Then, the driver assistance control unit 41 outputs a command to the vehicle control unit 42 to change the driving lane of vehicle 1 along the driving path 58.

[0226] If the instruction is accepted, the lane control unit 42 controls the drive control unit 43, the braking control unit 44, and the steering control unit 45 to make the vehicle 1 travel along the driving path 58, thereby changing the driving lane to the right lane of the current driving lane. The information output unit 17 outputs the coordinates of the lane boundary determined by the mapping matching unit 16, the type of driving lane (solid line and dashed line), and the vehicle position to the driver assistance control unit 41. The driver assistance control unit 41 compares the coordinates of the driving lane boundary, the type of driving lane (dashed line and solid line), and the vehicle position with the driving lane boundary lines obtained by the vehicle perimeter measurement unit 21 to control the distance between the vehicle position and the left and right white lines of the driving lane.

[0227] According to this structure, for example, even if the distance between vehicle 1 and other vehicles ahead decreases and the front stereo camera 22 is temporarily unable to detect the left and right white lines of the driving lane, as long as the prediction error of the vehicle's position remains below a predetermined error, it is possible to determine whether to change lanes using at least one of the types of white lines from the information output unit 17 and the types of white lines from the vehicle perimeter measurement unit 21. This enables driving assistance to ensure that vehicle 1 complies with traffic rules.

[0228] Furthermore, the driver assistance control unit 41 uses the coordinates of the lane boundary determined by the mapping matching unit 16 and the vehicle's position from the information output unit 17 to assist in lane keeping. Therefore, for example, even if the distance between the vehicle 1 and other vehicles ahead decreases, and the front stereo camera 22 is temporarily unable to detect the left and right white lines of the driving lane, as long as the prediction error of the vehicle's position remains below a predetermined error, stable lane keeping assistance can be provided to the driver and passengers.

[0229] Figures 47-51 This diagram illustrates an example of how vehicle 1 can overtake other vehicles 56d that are stopped ahead, following a chronological order.

[0230] Figure 47 In the middle, the millimeter-wave radar 23 at the front detects an obstacle in front of the vehicle 1's driving lane, and the stereo camera 22 at the front detects that the right side of the driving lane is a dashed line. The driver assistance control unit 41 maps the coordinates of the vehicle's position, the coordinates of the obstacle in front of the vehicle 1, and the type of white line to the virtual space in lane units.

[0231] Figure 48 The following scenario is illustrated: Vehicle 1 is approaching an obstacle until the stereo camera 22 detects that the obstacle is another parked vehicle 56d at a distance. In this case, if the operation input unit 19 receives an operation from the driver to implement the lane change assist function, the driver assistance control unit 41 generates a driving path 58 that extends beyond the other vehicle 56d and adds it to the virtual space, continuously updating the vehicle's position and the mapping of the lane lines, etc. In the driving path 58, a predetermined distance is set between vehicle 1 and the other vehicle 56d at the lateral position where vehicle 1 passes.

[0232] Figure 49 In the middle, the front stereo camera 22 confirms the positional relationship between the vehicle 1 and other vehicles 56d, and the vehicle 1 moves along a driving path 58 that avoids other vehicles 56d. The driver assistance control unit 41 controls the drive control unit 43, the braking control unit 44, and the steering control unit 45, and updates the information mapped to the virtual space so that the vehicle position and direction of travel calculated by the composite positioning unit 14 follow the driving path 58. In addition, this control and update also... Figure 50 and Figure 51 This will be carried out under certain circumstances.

[0233] Figure 50 The diagram shows the lateral movement of vehicle 1 as it passes other vehicles 56d. A predetermined distance d16 is provided between vehicle 1 and other vehicles 56d. If vehicle 1... Figure 50As the vehicle moves along the driving path 58, the millimeter-wave radar 25 on the left front gradually stops detecting other vehicles 56d, while the millimeter-wave radar 27 on the left rear gradually detects other vehicles 56d.

[0234] Figure 51 In the middle, the millimeter-wave radar 27 on the left rear cannot detect other vehicles 56d, and this vehicle 1 moves on the driving path 58 to return to the original lane.

[0235] According to this structure, for example, when there is another vehicle 56d parked in front of vehicle 1, the following driving assistance is provided: vehicle 1 moves along a driving path 58 with a distance d16 between vehicle 1 and other vehicle 56d while confirming that there are no other obstacles around vehicle 1. Therefore, vehicle 1 can avoid continuing to wait behind other vehicle 56d, thus suppressing traffic congestion.

[0236] Figures 52-54 The diagram illustrates the following situation: When vehicle 1 is traveling on a road with one lane on one side and approaches an intersection, it temporarily stops in front of the intersection and then proceeds forward towards the intersection.

[0237] Figure 52 In the middle, the front stereo camera 22 detects that there is a pedestrian crossing and a temporary stop line between the vehicle 1 and the intersection, and that the white lines on the left and right sides of the driving lane are solid lines. The driver assistance control unit 41 maps the vehicle's position, the coordinates and size of the white lines of the driving lane, the temporary stop line and the pedestrian crossing to, for example, a virtual space.

[0238] Figure 53 The image shows the vehicle 1 stopped at a temporary stop line near an intersection. In this situation, the front stereo camera 22 detects the pedestrian crossing ahead of the intersection, and the front millimeter-wave radar 23 detects obstacles in front of the pedestrian crossing. The driver assistance control unit 41 maps the coordinates and sizes of the pedestrian crossing and obstacles in front of the intersection. Furthermore, the driver assistance control unit 41 determines whether the distance d18 between the pedestrian crossing and the obstacle is greater than the length of the vehicle 1, i.e., whether there is space between the pedestrian crossing and the obstacle that the vehicle 1 can squeeze through. Figure 53 The diagram illustrates a situation where there is space between the pedestrian crossing and other vehicles 56e ahead, allowing the vehicle 1 to enter. In this case, the driver assistance control unit 41 determines that the vehicle 1 can proceed to the intersection when the traffic light turns green, and the display unit 18 notifies the driver and passengers of this decision via display and sound, thus enabling the vehicle 1 to enter the intersection.

[0239] Figure 54 In China, unlike Figure 53This indicates a situation where there is no space for vehicle 1 to enter between the pedestrian crossing and other vehicles 56e ahead. In this case, the driver assistance control unit 41 determines that vehicle 1 cannot proceed to the intersection when the traffic light turns green, and the display unit 18 informs the driver and passengers with the display of image 54 and sound that they should wait until there is space, and keeps vehicle 1 stationary.

[0240] According to this structure, vehicle 1 temporarily stops at the temporary stop line in front of the intersection, and then confirms whether there is space ahead of the intersection. Therefore, it can prevent situations where vehicle 1 forcibly enters the intersection when there is clearly no space ahead, thus causing traffic flow obstruction.

[0241] <Variation Example> In embodiment 7, the coordinates of other vehicles are mapped to a virtual space. However, the movement of other vehicles can also be predicted based on the coordinates of other vehicles at any given moment and the relative speed between other vehicles and vehicle 1. According to this structure, vehicle control that reflects the traffic conditions around vehicle 1 more accurately can be performed.

[0242] Furthermore, while Embodiment 7 describes the use of a stereo camera to detect the size of other vehicles, it is not limited to this. For example, the driver assistance control unit 41 can pre-associate vehicle models with their sizes in a database, and detect the size of other vehicles by identifying vehicle models whose outlines closely match the brightness boundaries detected by the stereo camera. Additionally, the driver assistance control unit 41 can calculate the distance between its own vehicle and other vehicles based on the sizes of other vehicles detected by the vehicle model and the sizes of other vehicles detected by the stereo camera.

[0243] <Other variations> The following will be the above Figure 1 The GNSS receiver 11, positioning enhancement signal receiver 12, positioning satellite selection unit 131, individual positioning solution calculation unit 132, floating-point solution calculation unit 133, ambiguity search and verification unit 134, fixed solution calculation unit 135, and satellite positioning error prediction unit 136 are denoted as "GNSS receiver 11, etc." The GNSS receiver 11, etc., consists of... Figure 55The processing circuit 81 shown implements this. Specifically, the processing circuit 81 includes: a GNSS receiver 11, which acquires observation data and orbit data from multiple GNSS satellites; a positioning augmentation signal receiver 12, which acquires positioning augmentation data from positioning augmentation satellites or the Internet; a positioning satellite selection unit 131, which selects a positioning satellite from multiple GNSS satellites; a separate positioning solution calculation unit 132, which calculates a separate positioning solution based on the observation data and orbit data of the positioning satellite without using positioning augmentation data; and a floating-point solution calculation unit 133, which calculates a solution based on the positioning satellite... The system uses observation data, orbit data, and positioning augmentation data to derive a floating-point solution including carrier phase offset; an ambiguity search and verification unit 134 derives an integer offset based on the carrier phase offset of the floating-point solution; a fixed solution calculation unit 135 derives a fixed solution based on satellite observation data, orbit data, positioning augmentation data, and integer offset; and a satellite positioning error prediction unit 136, which sets any one of the following as the positioning solution: a standalone positioning solution, a floating-point solution, a fixed solution, and a non-positioning solution indicating that the solution does not exist, and predicts the positioning error of the positioning solution for each epoch. The processing circuit 81 can be made of dedicated hardware or a processor that executes programs stored in memory. The processor can be, for example, a central processing unit, a processing device, an arithmetic unit, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor).

[0244] When the processing circuit 81 is dedicated hardware, it may be equivalent to a single circuit, a composite circuit, a programming processor, a parallel programming processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. The functions of each part of the GNSS receiver 11 can be implemented by circuits obtained by distributing the processing circuit, or the functions of each part can be concentrated into a single processing circuit.

[0245] When the processing circuit 81 is a processor, the functions of the GNSS receiver 11, etc., are implemented in combination with software, etc. Furthermore, the software, etc., may be, for example, software, firmware, or both software and firmware. The software, etc., is expressed in the form of a program and stored in memory. Figure 56As shown, the processor 82, which is used in the processing circuit 81, reads and executes the program stored in the memory 83, thereby realizing the functions of each part. That is, the positioning device has a memory 83 for storing a program, which, when executed by the processing circuit 81, ultimately performs the following steps: acquiring observation data and orbit data of multiple GNSS satellites; acquiring positioning augmentation data from a positioning augmentation satellite or the Internet; selecting a positioning satellite from multiple GNSS satellites; calculating a single positioning solution based on the observation data and orbit data of the positioning satellite without using positioning augmentation data; calculating a floating-point solution including carrier phase offset based on the observation data and orbit data of the positioning satellite and the positioning augmentation data; calculating an integer offset based on the carrier phase offset of the floating-point solution; calculating a fixed solution based on the observation data and orbit data of the positioning satellite, the positioning augmentation data, and the integer offset; and setting any one of the single positioning solution, the floating-point solution, the fixed solution, and a non-positioning solution indicating that the solution does not exist as the positioning solution, and predicting the positioning error of the positioning solution for each epoch. In other words, the program can also be said to enable the computer to perform the steps and methods of GNSS receiver 11, etc. Here, the memory 83 may be, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), as well as HDD (Hard Disk Drive), magnetic disk, floppy disk, optical disk, compressed disk, mini disk, DVD (Digital Versatile Disk) and its drive device, or any storage medium used thereafter.

[0246] The above describes the structure in which the functions of the GNSS receiver 11 and the like are implemented by either hardware or software. However, it is not limited to this; it is also possible to implement one part of the GNSS receiver 11 and the like using dedicated hardware and another part using software. For example, for the GNSS receiver 11, its functions can be implemented using a processing circuit 81, an interface, and a receiver, which are dedicated hardware. For other functional parts, their functions can be implemented by the processing circuit 81, which is a processor 82, reading the program stored in the memory 83 and executing it.

[0247] As described above, the processing circuit 81 can implement the above functions using hardware, software, or a combination thereof.

[0248] Furthermore, the positioning device 1 described above can also be applied to a positioning system that appropriately combines a PND (Portable Navigation Device), a navigation device, a DMS (Driver Monitoring System) or other vehicle device, a communication terminal including a mobile phone, smartphone, or tablet computer, the functions of an application installed in at least one of the vehicle device and the communication terminal, and a server to form a system. In this case, the various functions or structural elements of the positioning system described above can be distributed among the various devices that construct the system, or they can be centrally distributed in a single device.

[0249] Furthermore, it is possible to freely combine various implementation methods and variations, and to appropriately modify or omit various implementation methods and variations.

[0250] The above description is illustrative in all respects, not limiting. It can be understood that countless variations not shown can be conceived. Label Explanation

[0251] 11GNSS receiver 12 Positioning Enhancement Signal Receiver 131 Positioning uses satellite selection unit 132 Individually located solution calculation units 133 Floating-point solution calculation unit 134 fuzziness search verification units 135 Fixed Solution Calculation Unit 136 Satellite Positioning Error Prediction Unit 141 Speed ​​Sensor 144 angular velocity sensor 147 Autonomous Navigation Units 148 composite positioning units 149 Composite positioning error prediction unit.

Claims

1. A positioning device, characterized in that include: A GNSS acquisition unit acquires observation data, including pseudo-range, carrier phase, and Doppler offset frequency, as well as orbit data of the multiple GNSS satellites, according to each positioning signal from multiple GNSS satellites; A positioning augmentation data acquisition unit acquires positioning augmentation data from positioning augmentation satellites or the Internet; The positioning satellite selection unit selects a positioning satellite from a plurality of said GNSS satellites; A separate positioning solution calculation unit, which calculates a separate positioning solution based on the satellite observation data and the orbit data without using the positioning augmentation data; A floating-point solution calculation unit, which calculates a floating-point solution including carrier phase offset based on the observation data and orbit data of the positioning satellite and the positioning enhancement data. The search verification unit calculates the integer value offset based on the carrier phase offset of the floating-point solution; A fixed solution calculation unit calculates a fixed solution based on the observation data and orbit data of the positioning satellite, the positioning augmentation data, and the integer value offset. as well as A satellite positioning error prediction unit is configured to use any one of the following as a positioning solution: the individual positioning solution, the floating-point solution, the fixed solution, and a non-positioning solution indicating that the solution does not exist. This unit then predicts the positioning error of the positioning solution at each epoch. The satellite positioning error prediction unit predicts the positioning error of the floating-point solution at each epoch based on information from at least one of the following: the configuration of the positioning satellite, the time elapsed since the acquisition of the positioning augmentation data, the observation status of the positioning satellite, and the convergence status of the floating-point solution, and the relationship between these information and the positioning error of the floating-point solution. It also predicts the positioning error of the fixed solution at each epoch based on the relationship between this information and the positioning error of the fixed solution.

2. The positioning device as described in claim 1, characterized in that, The positioning signal also includes ionospheric delay error. The positioning signal includes a first positioning signal and a second positioning signal with different frequency bands. The positioning uses a satellite selection unit to calculate the first pseudo-range residual based on a comparison between the pseudo-range contained in the received positioning signal and the carrier phase or the Doppler offset frequency. When both the first and second positioning signals are available, the second pseudo-range residual is calculated based on the cancellation of the ionospheric delay errors of the first and second positioning signals. Based on at least one of the first pseudo-range residual and the second pseudo-range residual of the plurality of GNSS satellites, a temporary positioning satellite with communication quality above a first threshold is selected from the plurality of GNSS satellites. If the amount of observation data of the satellite used for temporary positioning exceeds a threshold, the positioning satellite is selected from the plurality of GNSS satellites based on at least one of the first pseudo-range residual and the second pseudo-range residual, wherein the communication quality is above a second threshold that is higher than the first threshold and the amount of observation data is below the threshold.

3. The positioning device as described in claim 1, characterized in that, Learn the relationship between the information and the positioning error of the floating-point solution, and the relationship between the information and the positioning error of the fixed solution.

4. The positioning device as described in claim 1, characterized in that, Under predetermined operational conditions, the floating-point solution calculation unit calculates the floating-point solution for each epoch, regardless of whether the integer bias is calculated by the search and verification unit. The search verification unit operates during the period from when the integer bias value is calculated until the radio waves of the positioning satellite are cut off or blocked, or from when the integer bias value is calculated until the positioning satellite is updated. If the positioning error of the fixed solution is greater than the first threshold and the pseudo-range residual of the positioning using satellites is smaller than the second threshold, the integer bias value is calculated.

5. The positioning device as described in claim 1, characterized in that, The satellite positioning error prediction unit, upon obtaining either the floating-point solution or the fixed solution, predicts the positioning error of the individual positioning solution based on the positioning error of the individual solution and the difference between the individual solution and the individual positioning solution.

6. The positioning device of claim 1, wherein, Also includes: An autonomous navigation unit that uses sensors to infer the vehicle's position; A composite positioning unit that calculates a composite positioning solution based on the vehicle's position inferred by the autonomous navigation unit and the individual positioning solutions; as well as A composite positioning error prediction unit, which, upon obtaining either the floating-point solution or the fixed solution, predicts the positioning error of the composite positioning solution based on the positioning error of the arbitrary solution and the difference between the arbitrary solution and the composite positioning solution.

7. The positioning device as described in claim 6, characterized in that, When neither the floating-point solution nor the fixed solution is found, the composite positioning unit uses the positioning error of the individual positioning solution obtained up to a predetermined epoch to correct the positioning error of the individual positioning solution, and then calculates the composite positioning solution based on the corrected individual positioning solution.

8. A positioning method, characterized in that, Observational data, including pseudo-range, carrier phase, and Doppler offset frequency, as well as orbital data of the multiple GNSS satellites, are acquired based on each positioning signal from multiple GNSS satellites. Obtain positioning augmentation data from positioning augmentation satellites or the Internet. Select a positioning satellite from among the multiple GNSS satellites mentioned. Without using the aforementioned positioning augmentation data, a separate positioning solution is derived based on the positioning using the satellite observation data and the orbital data. Based on the satellite observation data, orbit data, and positioning enhancement data used in the positioning process, a floating-point solution including carrier phase offset is obtained. The integer bias is calculated based on the carrier phase bias of the floating-point solution. A fixed solution is derived based on the satellite observation data and orbital data, the positioning augmentation data, and the integer bias. The localization solution is selected from the following: the single localization solution, the floating-point solution, the fixed solution, and the non-localization solution indicating that the solution does not exist. The localization error of the localization solution is then predicted for each epoch. Based on information from at least one of the following: the configuration of the positioning satellites used, the time elapsed since the acquisition of the positioning augmentation data, the observation status of the positioning satellites used, and the convergence status of the floating-point solution, and the relationship between these information and the positioning error of the floating-point solution, the positioning error of the floating-point solution is predicted for each epoch. Furthermore, based on the relationship between this information and the positioning error of the fixed solution, the positioning error of the fixed solution is predicted for each epoch.

9. A positioning device, characterized in that include: A GNSS acquisition unit acquires observation data, including pseudo-range, carrier phase, and Doppler offset frequency, as well as orbit data of the multiple GNSS satellites, according to each positioning signal from multiple GNSS satellites; A positioning augmentation data acquisition unit acquires positioning augmentation data from positioning augmentation satellites or the Internet; The positioning satellite selection unit selects a positioning satellite from a plurality of said GNSS satellites; A separate positioning solution calculation unit, which calculates a separate positioning solution based on the satellite observation data and the orbit data without using the positioning augmentation data; A floating-point solution calculation unit, which calculates a floating-point solution including carrier phase offset based on the observation data and orbit data of the positioning satellite and the positioning enhancement data. The search verification unit calculates the integer value offset based on the carrier phase offset of the floating-point solution; A fixed solution calculation unit calculates a fixed solution based on the observation data and orbit data of the positioning satellite, the positioning augmentation data, and the integer value offset. A satellite positioning error prediction unit is configured to use any one of the following as a positioning solution: the individual positioning solution, the floating-point solution, the fixed solution, and a non-positioning solution indicating that the solution does not exist. This unit then predicts the positioning error of the positioning solution at each epoch. An autonomous navigation unit that uses sensors to infer the vehicle's position; as well as A composite positioning unit, which calculates a composite positioning solution based on the vehicle's position inferred by the autonomous navigation unit and the positioning solution. The satellite positioning error prediction unit predicts the positioning error of the floating-point solution at each epoch based on at least one of the following: the configuration of the positioning satellites used, the time elapsed since the acquisition of the positioning augmentation data, the observation status of the positioning satellites used, and the convergence status of the floating-point solution, and the relationship between these factors and the positioning error of the floating-point solution. Furthermore, based on the relationship between this information and the positioning error of the fixed solution, it predicts the positioning error of the fixed solution at each epoch. It also includes a control unit that, based on the composite positioning solution and map data, calculates at least one of the following distances: the distance between the center of the vehicle and the center of the vehicle's driving lane, the distance between the vehicle and the white line on the left side of the driving lane, and the distance between the vehicle and the white line on the right side of the driving lane, and causes the display unit to display a color corresponding to the at least one distance.

10. The positioning device as described in claim 9, characterized in that, Based on the composite positioning solution and the map data, the control unit causes the display unit to display the vehicle's driving trajectory and the position of the white line in the driving lane, and causes the display unit to perform attention activation corresponding to the driving trajectory and the position of the white line.

11. The positioning device as described in claim 9, characterized in that, Based on the composite positioning solution and the map data, the control unit causes the display unit to display the boundary line of the driving lane and the positional relationship between the white line and the vehicle.

12. The positioning device as described in claim 9, characterized in that, Based on the composite positioning solution and the map data, the control unit directs the display unit to the emergency stopping lane in front of the vehicle.

13. The positioning device as described in claim 9, characterized in that, Based on the composite positioning solution and the map data, the control unit causes the display unit to display the destination aspects of each of the multiple lanes containing the driving lane.

14. The positioning device as described in claim 9, characterized in that, Based on the composite positioning solution and the map data, the control unit causes the display unit to perform attention activation when the vehicle enters or has entered a lane that the vehicle cannot pass through.

15. The positioning device as described in claim 9, characterized in that, It also includes a measuring unit that detects the left and right white lines of the driving lane and, based on the combination of detected white line types, determines which of the following lanes the vehicle is traveling in: the left lane, the right lane, or the inner lane. The control unit, based on the composite positioning solution and the estimation results in the measurement unit, causes the display unit to guide the vehicle to the lane it should travel in.

16. The positioning device as described in claim 15, characterized in that, The measuring unit detects protrusions and depressions in the road surface in front of the vehicle, as well as obstacles in front of the vehicle. The control unit causes the display unit to perform attention arousal based on the detection results of the obstacle.

17. The positioning device as described in claim 15, characterized in that, The measuring unit detects obstacles in front of the vehicle. Based on the map data and the detection results of the obstacles, the control unit causes the display unit to display a map that maps the lane units of the obstacles.

18. The positioning device as described in claim 15, characterized in that, The measuring unit uses multiple detection devices to detect obstacles on the sides of the vehicle. The control unit predicts the behavior of the obstacle based on the detection results of the obstacle.

19. The positioning device as described in claim 15, characterized in that, The measuring unit determines the road surface condition on which the vehicle is traveling. The control unit causes the display unit to perform attention activation based on the determination result of the road surface condition.

20. The positioning device as described in claim 15, characterized in that, The control unit controls the distance between the vehicle and the white line based on the composite positioning solution and the detection result of the white line.

21. The positioning device as described in claim 15, characterized in that, The control unit controls the change of the vehicle's driving lane based on the composite positioning solution, the detection result of the white line, and the type of the detected white line.

22. The positioning device as described in claim 15, characterized in that, The measuring unit detects parked vehicles in front of the vehicle. Based on the composite positioning solution and the detection results of the parked vehicle, the control unit enables the vehicle to overtake the parked vehicle.

23. The positioning device as described in claim 15, characterized in that, The measuring unit detects the portion of the road ahead of the vehicle at the intersection. Based on the detection results of the measuring unit, if the control unit determines that there is space in the part that the vehicle can enter, it causes the vehicle to enter the part.

Images