JP4738196B2 - Method for determining error circle of position calculation device - Google Patents

Description

  The present invention relates to a position calculation device that calculates a predicted position of a moving object using positioning results from a GPS (Global Positioning System) satellite and a self-contained navigation sensor, and in particular, determination of an error circle indicating the positioning accuracy of a GPS position by GPS positioning. Regarding the method.

  Methods for calculating the position of the vehicle include GPS positioning using a GPS satellite and autonomous navigation positioning using a gyro sensor or a vehicle speed sensor mounted on the vehicle. GPS positioning can detect the absolute position and absolute azimuth, but has the disadvantage that its accuracy deteriorates depending on the reception environment. On the other hand, self-contained navigation positioning does not depend on the reception environment, but has a disadvantage that errors accumulate. For this reason, in navigation devices, a hybrid system using positioning data from a GPS satellite and positioning data from a self-contained navigation sensor is widely used so as to complement both disadvantages.

  In Patent Document 1, a positioning error ΔLg of a GPS satellite is determined based on a degradation coefficient of GDOP (Geometrical Dilution Of Precision) which is a positioning environment between the GPS satellite and a receiver, and is compared with a positioning error D by a self-contained navigation. In the case of D> ΔLg, a navigation device is disclosed that determines the position by GPS positioning as the current position, and in the case of D ≦ ΔLg, determines the positioning position D by self-contained navigation as the current position.

  Patent Document 2 relates to a vehicle position correction device, and compares a self-contained navigation travel distance of a section obtained by self-contained navigation with a distance between GPS positioning points of a section corresponding to the certain section obtained by GPS positioning. When the difference between the navigation travel distance and the distance between GPS positioning points is large, it is determined that the reliability of the GPS positioning data is low, and when the difference is small, it is determined that the reliability is high. If the reliability is high, the radius of the error circle indicating the allowable range for the GPS positioning error is reduced, and if the reliability is low, the radius is increased.

JP 63247612 A JP 2003-279362 A

  When performing position calculation using GPS positioning and self-contained navigation positioning as shown in Patent Document 2, in a case where the positioning environment is good, such as open air, the error circle radius due to the travel distance ratio is approximately correctly weighted. Although it can be determined, in the case of being affected by multipaths such as a shielded environment, GPS positions with large errors may be output continuously and may move in the same manner as the locus of autonomous navigation positioning. As a result, it is determined that the travel distance ratio is good, and although the GPS position error is large, weighting is performed and the accuracy of the predicted position deteriorates.

  For example, as shown in FIG. 4B, the navigation position L2 by the self-contained navigation positioning at the positioning time T2 is obtained by adding the velocity vector Vc to the predicted position Q1 at the positioning time T1. The speed vector Vc is determined from the relative direction of the own vehicle detected by the angle sensor and the direction speed of the own vehicle detected by the speed sensor. G1, G2,... Gn are absolute positions (hereinafter referred to as GPS positions) obtained by GPS positioning at positioning times T1, T2,. The travel distance ratio is a ratio of the sum of the distances between the two points of the GPS position and the sum of the magnitudes of the speed vectors, as shown in Formula 1. The section or period for calculating the travel distance is constant, and when GPS positioning is performed every second, the section is, for example, four sections (or four seconds).

  When the GPS positioning environment deteriorates due to the influence of multipath, satellite arrangement, etc., the GPS position jumps greatly, for example, from G3 to G4. After that, if a bad positioning state by GPS continues for a certain period, the GPS position becomes a linear locus such as G5, G6, G7, and G8. As described above, if the period for calculating the travel distance is 4 sections (4 seconds), the error circle of the GPS position G8 at the positioning time T8 is GPS positions G4-G5, G5-G6, G6-G7, G7- It is calculated from the ratio of the sum of the travel distances of the four sections Tm of G8 and the sum of the magnitudes of the speed vectors of the sections Tn corresponding thereto. In this case, the travel distance ratio between the two becomes close to “1”, and the radius of the error circle R becomes small although the positioning error of the GPS position is large. As a result, the navigation position L8 obtained by adding the velocity vectors is weighted to the GPS error circle, the position is determined as the predicted position Q8, and the accuracy of the predicted position Q8 is deteriorated. When the accuracy of the predicted position deteriorates, the map matching accuracy in the navigation device also deteriorates.

  An object of the present invention is to solve the above-described conventional problems, and to provide a position calculation device that can re-determine the amount of error when the positioning status by GPS deteriorates and improve the accuracy of a predicted position. A further object of the present invention is to provide a navigation device that can accurately display the position of the vehicle on a map using the calculated predicted position.

  The position calculation device according to the present invention calculates a predicted position of a moving body using a positioning result by GPS and a positioning result by a self-contained navigation sensor, and is independent of a travel distance Dp of a certain period Tc measured by GPS. An error amount determining means for determining an error amount R of the GPS position by GPS positioning based on a traveling distance ratio with the traveling distance Dq corresponding to the predetermined period Tc measured by the navigation sensor, and a navigation position measured by the self-contained navigation sensor A determination unit that determines whether or not L is within the error amount R of the GPS position, and when it is determined that the navigation position L does not exist in the error amount R, the travel distance Dp in the error amount determination unit Error amount redetermining means for extending the predetermined period Tc for calculating the travel distance Dq by a predetermined period Ts and redetermining the error amount R.

  Preferably, the determination unit compares the error amount R with the navigation position L, and when it is determined that L> R, the error amount re-determination unit sets the travel distance ratio in the extended period (Tc + Ts). Based on this, the error amount R is redetermined.

  The position calculation device further determines the navigation position L when the navigation position L exists in the error amount R after the error amount is redetermined, and the navigation position L exists in the error amount R. When not, a position calculating means for using the GPS position as a predicted position is included.

  The position calculation method according to the present invention is based on the GPS distance measurement based on the travel distance ratio between the travel distance Dp of the fixed period Tc measured by GPS and the travel distance Dq corresponding to the fixed period Tc measured by the self-contained navigation sensor. A step of determining a position error amount R and, when it is determined that the navigation position measured by the self-contained navigation sensor does not exist within the error amount R of the GPS position, for calculating the travel distance Dp and the travel distance Dq Extending the predetermined period Tc by a predetermined period Ts and re-determining the error amount R; and after re-determining the error amount R, when the navigation position L exists within the error amount R, A navigation position L as a predicted position, and when the navigation position L does not exist within the error amount R, the GPS position is set as a predicted position.

  According to the present invention, when the navigation position measured by the self-contained navigation sensor is not included in the GPS position error amount R, the period for calculating the travel distance is made longer than usual, and the error is calculated. Since the amount R is re-determined, the error amount reflecting the positioning error of the GPS position can be determined, and thereby the accuracy of the predicted position calculated can be improved.

  The best mode for carrying out the present invention will be described below in detail with reference to the drawings. The position calculation device according to the present invention is preferably used in an in-vehicle navigation device.

  FIG. 1 is a block diagram illustrating a configuration of a position calculation apparatus according to an embodiment of the present invention. The position calculation device 1 receives a GPS signal sent from a GPS satellite, outputs a positioning data related to the absolute position of the vehicle, the absolute direction using the Doppler shift, and the GPS speed, and is attached to the vehicle. The self-contained navigation sensor 20 that outputs positioning data related to the relative azimuth and azimuth speed by the various sensors, and the positioning data output from the GPS receiver 10 and the self-contained navigation sensor 20 are input, and the moving object is detected based on the input positioning data A position calculation unit 30 that calculates a predicted position and a memory 40 that stores travel history information and the like of the moving body are included.

The GPS receiver 10 includes a GPS receiving antenna 12, a receiving unit 14, and a calculating unit 16. The calculation unit 16 calculates the GPS position P _GPS , GPS azimuth θ _GPS , and GPS speed S _GPS at the positioning time per second, and outputs these positioning data to the position calculation unit 30.

The self-contained navigation sensor 20 includes an angular velocity sensor 22 such as a gyro that detects the rotation angle of the vehicle, and the angular velocity sensor 22 detects the navigation direction θ _GYRO (relative direction) and outputs it. The self-contained navigation sensor 20 further includes a speed sensor 24 that generates a pulse each time the moving body travels a certain distance, and the speed sensor 24 outputs an azimuth speed V.

  The position calculation unit 30 is configured by, for example, a microcomputer, and executes calculation of a predicted position by a program stored in a ROM / RAM. The memory 40 sequentially stores, as history information, the predicted position calculated by the position calculation unit 30, past positioning data by GPS, positioning data by self-contained navigation, and the like.

Next, the operation of the position calculation unit 30 will be described. First, FIG. 2 shows a flow for determining an error circle that is the positioning accuracy of the GPS position. The position calculation unit 30 receives the positioning data from the GPS receiver 10 and the positioning data from the self-contained navigation sensor 20 (step S101). The position calculation unit 30 calculates the travel distance Dp from the sum of the distances between two GPS positions over a certain positioning period, and calculates the travel distance Dq measured by the self-contained navigation sensor during the period corresponding to the certain positioning period. Calculate (step S102). The travel distance Dq is the sum of the magnitudes of the speed vectors, and the speed vector is defined by the navigation direction θ _GYRO of the navigation direction and the direction speed V. The fixed positioning period in the normal time is, for example, 4 seconds (or 4 sections).

  Next, the position calculation unit 30 calculates a travel distance ratio (Dp / Dq) (step S103), and determines whether the travel distance ratio belongs to a range of h1 <(Dp / Dq) <h2 ( Step S104). This range is, for example, 0.9 <(Dp / Dq) <1.1. The closer the mileage ratio is to 1, the higher the reliability of GPS positioning. Therefore, the position calculation unit 30 sets an error circle with a radius r1 when the travel distance ratio is within the above range (step S105), and sets an error calculation circle with a radius r2 when it is out of the range (step S106). ). The radius is in a relationship of r1 <r2. Here, in order to simplify the explanation, the size of the error circle is set in two stages, but of course, it may be set in detail in three or more stages.

Next, the calculation operation of the predicted position using the positioning data by GPS and the positioning data by self-contained navigation will be described with reference to the flow of FIG . First, the navigation position L is obtained by adding the velocity vector to the previously calculated predicted position (step S201). Next, the position control unit 30 obtains an error circle R of the GPS position measured this time according to the operation flow shown in FIG. 2 based on the positioning data by GPS and the positioning data by self-contained navigation (step S202). The position control unit 30 compares the error circle R and the navigation position L (step S203), and determines whether the navigation position L exists in the error circle R (step S204).

  If the navigation position L is within the range of the error circle R, the navigation position L becomes the current predicted position (step S208). On the other hand, when the navigation position L does not exist within the range of the error circle R, the position control unit 30 recalculates the error circle (step S205).

  In the recalculation of the error circle, the positioning period for calculating the travel distance Dp by GPS positioning and the travel distance Dq by self-contained navigation is extended by a predetermined period. In this embodiment, the normal positioning period is 4 seconds. However, when the error circle is recalculated, the positioning period is extended to at least 10 seconds. Referring to past actual measurement values, the period in which positioning is degraded due to the influence of multipath or the like hardly lasts for 10 seconds or more. For this reason, if the positioning period for calculating the travel distance ratio is 10 seconds, even if a part of the travel distance ratio in the positioning period is close to 1, the overall travel distance ratio reflects the deterioration of the positioning environment. As a result, the radius of the error circle increases.

  The position calculation unit 30 compares the recalculated error circle R1 with the navigation position L again (step 206), and if the navigation position L is within the range of the error circle R1 (step S207), the navigation position. L is the predicted position (step S208). If the navigation position L does not exist in the error circle R1, the navigation position L is attracted to the center of the error circle, that is, the GPS position, and this is set as the predicted position (step S209).

If navigation position L> error circle R in step S204, the velocity vector position is outside the error circle radius, and the following two cases are conceivable.
(1) The GPS positioning state is poor, and GPS positions with large errors are output continuously for a certain period of time.
(2) Positioning by GPS is very good and the radius of error circle is very small.
In the case of (1), by checking the long-distance travel distance ratio, the stability of the trajectory can be strictly determined and the correct weighting amount can be determined.
In the case of (2), since the original positioning state is good, the long-distance travel distance ratio remains good and there is almost no influence.

  FIG. 4A is a diagram illustrating a calculation example of a predicted position by the position calculation apparatus according to the present embodiment. The predicted position at the positioning time T1 is Q1. The navigation position L2 at the positioning time T2 is obtained by adding the velocity vector Vc at the positioning time T1 to the predicted position Q1. G1, G2,... Gn are GPS positions at positioning times T1, T2,. Since the navigation position K2 at the positioning time T2 is within the range of the error circle R of the GPS position G2, the navigation position L2 is determined as the predicted position Q2 at the positioning time T2. Next, the navigation position L3 at the positioning time T3 is obtained by adding the velocity vector Vc to the predicted position Q2. The GPS positioning environment is slightly deteriorated at the positioning time T3, and the GPS position G3 fluctuates. However, since the navigation position L3 is within the error circle R of the GPS position G3, the navigation position L3 becomes the predicted position Q3. It is determined.

At the positioning time T4, the navigation position L4 is calculated based on the predicted position Q3 in the same manner as described above. Further, at the positioning time T4, the GPS positioning environment is further deteriorated, and the GPS position G4 jumps greatly. Since the GPS position G3 and between two points between G4 distance K ₃₄ becomes very large, the travel distance ratio deviates more than 1, is set large radius error yen R1 indicating that positioning errors of the GPS position is greater . The navigation position L4 is compared with the error circle R1 of the GPS position G4. Since the navigation position L4 is within the range of the error circle R1, the navigation position L4 becomes the predicted position Q4.

Thereafter, until the positioning time T7, since the distance traveled by the GPS positioning contains a G3 and between two points of the G4 distance _{K 34} is a GPS position, a large error yen R1 radius is continued, navigation position L5, L6, L7 is determined as the predicted positions Q5, Q6, and Q7, respectively.

  At the positioning time T8, the distance traveled by GPS positioning is calculated from four sections G4-G5, G5-G6, G6-G7, and G7-G8, which are distances between two points. The GPS position trajectory of this section approximates the velocity vector trajectory of the corresponding positioning period. For this reason, the travel distance ratio becomes close to “1”, and the position calculation unit 30 determines that the error due to GPS positioning is small, and sets an error circle R having a small radius. Thereby, the navigation position L8 exists outside the range of the error circle R. In the case of the conventional method, the navigation position L8 is attracted to the error circle R. In this embodiment, the error circle is recalculated as described above, and then the navigation position L8 and the error circle are recompared. Do.

In recalculation of the error circle, to a positioning period 10 seconds (or 10 positioning period), the travel distance Dp by GPS positioning, since that will contain the distance _{K 34} of G3-G4, running distance ratio is from 1 Also deviate. Accordingly, the error circle at the GPS position G8 at the positioning time T8 is determined to be an error circle R1 having a large radius, as at the positioning time T7. As a result, the navigation position L8 exists within the range of the error circle R1, and the navigation position L8 is determined as the predicted position Q8. Therefore, the predicted position Q8 is not attracted to the GPS position G8 that includes a large amount of error, and the accuracy of the predicted position can be maintained.

  In this way, in the conventional position calculation method, even if the GPS positioning error is large, if the mileage ratio for a certain period is good, the error circle radius becomes small, so that the error in the predicted position is attracted to it. However, in this embodiment, when the navigation position and the GPS position are far apart (navigation position> error circle radius), the positioning period of the traveling distances Dp and Dq is extended and the quality of the traveling distance ratio is checked again. Since the determination is made, when the GPS positioning error is large, an error circle reflecting the error can be set, the effect of the weight is reduced, and the error of the predicted position does not increase. In other words, even when GPS positions with large errors are output continuously in the case of being affected by multipath or the like, a highly accurate predicted position can be calculated by determining an appropriate weighting amount.

  FIG. 5 is a block diagram when the position calculation apparatus according to the present embodiment is applied to a navigation apparatus. The navigation device 100 includes a GPS receiver 10 that receives radio waves from the GPS satellite described above, a self-contained navigation sensor 20, a VICS / FM multiplex receiver 116 that receives current road traffic information outside the vehicle via an antenna 114, and an operation panel. 122, user input interface 120 including voice input unit 124 and remote control operation unit 126, storage device 130 having a large-capacity hard disk, data communication control unit 132 enabling wireless or wired data communication, and output of sound from speaker 142 A display control unit 150 that displays an image on the display 152, a program memory 160 that stores various programs, a data memory 170 that temporarily stores data, and a control unit 180. The control unit 180 includes the error circle determination method and the predicted position calculation function of the position calculation unit 30 shown in FIG.

  The storage device 130 stores a program and a database for executing various navigation functions. The database includes map data and facility data, and the map data includes road link data and intersection data. The program memory 160 loads a program stored in the storage device 130, and includes a program for determining an error circle, a program for calculating a predicted position, and a predicted position calculated by GPS positioning and navigation positioning. A program for map matching the vehicle position on the map data, a program for searching for the optimum route to the destination, and the like are stored. The data memory 170 stores map data read from the storage device 130, positioning data from the GPS receiver 10 and the self-contained navigation sensor 20, and the like.

  Since a highly accurate predicted position can be obtained by the position calculation apparatus of the present embodiment, the navigation apparatus 100 can also reduce the processing burden due to map matching and display the vehicle position more quickly and accurately. it can.

  The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific embodiments according to the present invention, and various modifications can be made within the scope of the gist of the present invention described in the claims. Deformation / change is possible.

  The position calculation device according to the present invention is used in a navigation device or a navigation system for a moving body such as a vehicle.

It is a block diagram which shows the structure of the position calculation apparatus which concerns on the Example of this invention. It is a figure which shows the calculation flow of the error circle in the position calculation apparatus of a present Example. It is a figure which shows the calculation flow of the estimated position of the position calculation apparatus of a present Example. It is a figure explaining the example of calculation of the predicted position by the position calculation apparatus of a present Example. It is a figure which shows the structure of the navigation apparatus to which the position calculation apparatus of a present Example is applied.

Explanation of symbols

1: Position calculation device 10: GPS receiver 12: Antenna 14: Reception unit 16: Calculation unit 20: Self-contained navigation sensor 22: Angle sensor 24: Speed sensor 30: Position calculation unit 40: Memory

Claims (6)

A position calculation device that calculates a predicted position of a moving object using a positioning result by GPS and a positioning result by a self-contained navigation sensor,
An error for determining an error amount R of the GPS position by GPS positioning based on a traveling distance ratio between a traveling distance Dp measured by GPS for a certain period Tc and a traveling distance Dq corresponding to the certain period Tc measured by a self-contained navigation sensor. A quantity determining means;
Determining means for determining whether or not the navigation position L measured by the self-contained navigation sensor is within the error amount R of the GPS position;
When it is determined that the navigation position L does not exist within the error amount R , the predetermined period Tc for calculating the travel distance Dp and the travel distance Dq in the error amount determination means is extended by a predetermined period Ts. Error amount re-determination means for re-determining the error amount R ,
The period Ts is a positioning period that is earlier than the period Tc continuous to the period Tc, and the error amount re-determination means sets the travel distance ratio between the travel distance Dp and the travel distance Dq in the period (Tc + Ts). The error amount R is re-determined based on
Position calculation device. The position calculation device according to claim 1, wherein the travel distance Dq is a sum of magnitudes of velocity vectors obtained from a relative azimuth measured by an angle sensor and an azimuth velocity measured by a speed sensor. The position calculation device according to claim 1, wherein the navigation position L is calculated by adding a velocity vector to the previously calculated predicted position. The position calculation device further determines the navigation position L when the navigation position L exists in the error amount R after the error amount is redetermined, and the navigation position L exists in the error amount R. The position calculation device according to any one of claims 1 to 3 , further comprising position calculation means that uses the GPS position as a predicted position when not. A position calculation device according to any one of claims 1 to 4 ,
A navigation device comprising display means for displaying a vehicle position on a road map based on a predicted position calculated by a position calculation device. A position calculation method in a position calculation device that calculates a predicted position of a moving object using a positioning result by GPS and a positioning result by a self-contained navigation sensor,
A step of determining a GPS position error amount R by GPS positioning based on a traveling distance ratio between a traveling distance Dp measured by GPS for a certain period Tc and a traveling distance Dq corresponding to the certain period Tc measured by a self-contained navigation sensor. When,
When it is determined that the navigation position measured by the self-contained navigation sensor does not exist within the error amount R of the GPS position, the predetermined period Tc for calculating the travel distance Dp and the travel distance Dq is set to a predetermined period Ts. And re-determining the error amount R,
After re-determining the error amount R, when the navigation position L exists in the error amount R, the navigation position L is set as a predicted position, and when the navigation position L does not exist in the error amount R, the GPS A position as a predicted position ,
The period Ts is a positioning period that is earlier than the period Tc continuous to the period Tc, and the error amount re-determination means sets the travel distance ratio between the travel distance Dp and the travel distance Dq in the period (Tc + Ts) A position calculation method for re-determining the error amount R based on the position.

Images