JP6569572B2 - Vehicle position determination device - Google Patents

Description

  The present invention relates to a vehicle position determination device that determines the position of a vehicle using a GNSS receiver.

  A technique for determining the position of a vehicle using a GNSS receiver is widely known. The position detected by the GNSS receiver is the position where the antenna of the receiver exists. Therefore, when the mounting position of the antenna of the GNSS receiver on the vehicle is changed, the position detected by the GNSS receiver changes even if the position of the vehicle is the same.

  Therefore, in Patent Document 1, the relative position of the GPS receiver with respect to the vehicle body is specified by comparing the positioning result measured by the GPS receiver, which is a GNSS receiver, with the vehicle position specified by the roadside device through image recognition, and , Seeking the creditworthiness of its relative position. Then, the location information use service provided to the user is selected according to the reliability.

JP 2008-249666 A

  As described above, in Patent Document 1, the reliability of the relative position of the GPS receiver to the vehicle body is obtained. Then, the position information use service provided to the user is selected according to the reliability. Thereby, it is said that an appropriate service can be provided according to positioning accuracy.

  However, the positioning accuracy affects not only the reliability of the relative position of the GPS receiver to the vehicle body but also the reliability of the positioning results of a GNSS receiver such as a GPS receiver. Therefore, the reliability of Patent Document 1 has not sufficiently reflected the positioning accuracy.

  Patent Document 1 describes that the relative position of the GPS receiver with respect to the vehicle body is calculated as installation position data, and the position of the vehicle is calculated from the installation position data and the coordinates of the receiver. However, there is no description on how to specifically calculate the position of the vehicle.

  Here, it is assumed that the difference between the coordinates of the receiver and the vehicle coordinates when the installation position data is determined is (Δx, Δy). If the position of the vehicle is simply calculated by adding or subtracting (Δx, Δy) to the coordinates detected by the GPS receiver, the direction of the vehicle when the installation position data is determined and the position of the vehicle are calculated. If the direction of the vehicle is different, the position of the vehicle with high accuracy cannot be calculated.

The present invention has been made based on this situation, and determines the vehicle position, the purpose of providing a vehicle position determination apparatus capable of determining accurately the reliability of its position.

  The above object is achieved by a combination of the features described in the independent claims, and the subclaims define further advantageous embodiments of the invention. Reference numerals in parentheses described in the claims indicate a correspondence relationship with specific means described in the embodiments described later as one aspect, and do not limit the technical scope of the present invention. .

The invention for achieving the above object is a vehicle position determination device for determining the position of a vehicle, which is used in a vehicle equipped with a GNSS receiver that receives a navigation signal transmitted from a navigation satellite by an antenna, Based on the distance between the positioning position acquisition unit (S3) that sequentially calculates the satellite positioning absolute position, which is calculated based on the position of the antenna, and the distance between the object and the known coordinate determined using the distance sensor. A sensor positioning position determination unit (S6) for determining a sensor positioning absolute position that represents a determined vehicle position in absolute coordinates, and an antenna for a reference position on the vehicle based on a position difference between the satellite positioning absolute position and the sensor positioning absolute position An antenna position determining unit (S13) that determines an antenna position offset representing the position of the satellite, and correcting by correcting the satellite positioning absolute position based on the antenna position offset After correction for determining the positioning absolute position locator and (S33, S34), it calculates the antenna position prediction error is a value obtained by the prediction error of the antenna position offsets antenna position determination unit has determined, based on the antenna position prediction error An antenna position prediction error calculating unit (73) for determining the best antenna position prediction error, a satellite positioning prediction error acquiring unit (S42) for acquiring a satellite positioning prediction error that is a value obtained by predicting the error of the satellite positioning absolute position, A positioning reliability calculation unit (S45) that calculates a positioning reliability that is a reliability of the corrected absolute position of the positioning based on the best antenna position prediction error and the satellite positioning prediction error.

  In the present invention, a sensor positioning absolute position, which is a vehicle position determined based on a distance between an object with known coordinates and the vehicle, is determined, and an antenna is based on a position difference between the sensor positioning absolute position and the satellite positioning absolute position. The position offset is determined. Since the satellite positioning absolute position is corrected based on the antenna position offset, it is possible to determine a corrected positioning absolute position that is not easily affected by the change in the antenna position. This corrected positioning absolute position represents the vehicle position.

  In addition, an antenna position prediction error that is a value obtained by predicting an error in the antenna position offset is calculated, and a satellite positioning prediction error that is a value obtained by predicting an error in the satellite positioning absolute position is acquired. Based on the antenna position prediction error and the satellite positioning prediction error, the positioning reliability that is the reliability of the corrected positioning absolute position is calculated.

  In addition to the detection error of the antenna position, the error of the satellite positioning absolute position also affects the accuracy of the corrected positioning absolute position. Therefore, by calculating the positioning reliability based on the antenna position prediction error and the satellite positioning prediction error, it is possible to obtain a highly accurate positioning reliability.

It is a block diagram which shows the structure of the vehicle-mounted apparatus 1 provided with the function as a vehicle position determination apparatus with which this invention was applied. It is a figure explaining distance dx, dy which the horizontal direction detection sensor 20 of FIG. 1 and the advancing direction detection sensor 30 detect. It is a flowchart which shows the process which the matching part 71 of FIG. 1 performs. It is a flowchart which shows the process which the antenna position offset determination part 72 of FIG. 1 performs. It is a flowchart which shows the process which the antenna position prediction error calculation part 73 of FIG. 1 performs. It is a figure explaining the offset prediction error at the time of normalization computed by S26 of FIG. It is a flowchart which shows the process which the positioning position correction | amendment part 74 of FIG. 1 performs. It is a flowchart which shows the process which the positioning reliability output part 75 of FIG. 1 performs.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. As shown in FIG. 1, the in-vehicle device 1 of this embodiment having a function as a vehicle position determination device includes a GNSS receiver 10, a lateral direction detection sensor 20, a traveling direction detection sensor 30, a storage unit 40, and inertial navigation. The apparatus 50, the vehicle control part 60, and the calculating part 70 are provided. This in-vehicle device 1 is mounted on a vehicle 2 shown in FIG.

  The GNSS receiver 10 includes an antenna 11 that receives a navigation signal transmitted by a navigation satellite. The navigation satellite is a navigation satellite included in one or more of various satellite positioning systems such as GPS, GLONASS, Galileo, IRNSS, QZSS, and Beidou.

  The GNSS receiver 10 analyzes the navigation signal received by the antenna 11 to obtain a code pseudorange, a carrier phase, a Doppler frequency, and the like, and from these, absolute coordinates representing the current position of the antenna 11 (hereinafter referred to as satellite positioning absolute position). Are calculated sequentially. The satellite positioning absolute position is represented by latitude and longitude altitude. As a method of calculating the satellite positioning absolute position from the navigation signal, various known methods such as RTK-GNSS and DGNSS can be used in addition to a general method of positioning using only the navigation signal.

  The error of the satellite positioning absolute position calculated by the GNSS receiver 10 is dynamically changed because it is affected by the amount of reinforcement information, the satellite arrangement, the surrounding building environment, and the like. In addition, the reinforcement information means the information used for positioning other than a navigation signal, for example, the difference information in DGNSS corresponds to reinforcement information.

  In addition to sequentially calculating the satellite positioning absolute position, the GNSS receiver 10 outputs a satellite positioning prediction error that is a value representing the positioning accuracy. The positioning accuracy can be determined based on the satellite arrangement, the number of observable satellites, the amount of reinforcement information, and the like. Various known methods can be used as a method for determining the satellite positioning prediction error. The cycle in which the GNSS receiver 10 calculates the satellite positioning absolute position and the satellite positioning prediction error is, for example, 100 msec.

  The lateral direction detection sensor 20 detects an object outside the vehicle that exists in the lateral direction of the vehicle 2, that is, the vehicle width direction, and detects a distance in the lateral direction of the vehicle (hereinafter referred to as a lateral distance) dx to the object. This object includes not only a three-dimensional object but also a planar object such as the road marking line 3 shown in FIG. As the lateral direction detection sensor 20, for example, a white line recognition camera or a side radar can be used. The lateral direction detection sensor 20 also detects the shape of an object that exists in the lateral direction of the vehicle. The shape is detected by, for example, detecting distances to a plurality of points on the object. Both the lateral direction detection sensor 20 and the traveling direction detection sensor 30 described below correspond to the distance sensor in the claims.

  The lateral direction detection sensor 20 further outputs a lateral direction prediction error that is a value obtained by predicting the error of the lateral direction distance dx. For example, if the lateral direction detection sensor 20 is a camera, the lower the contrast, the lower the accuracy of the detection distance. Further, if the lateral direction detection sensor 20 is a radar, the accuracy of the detection distance decreases if the SN ratio is small. Therefore, there is a relationship between the reliability of the signal used to detect the object and the error in the detection distance. Therefore, the horizontal direction detection sensor 20 determines the horizontal direction prediction error based on the signal used to detect the object and the basic detection accuracy of the horizontal direction detection sensor 20, and outputs the horizontal direction prediction error. To do. This horizontal direction prediction error indicates a range including, for example, a measurement value of 95%, and is specifically expressed as ± 10 cm. The basic detection accuracy is, for example, detection accuracy on a catalog, and is set in advance.

  The traveling direction detection sensor 30 detects a distance (hereinafter referred to as a traveling direction distance) dy in the vehicle traveling direction to an object existing in front of the traveling direction of the vehicle 2. The object detected by the traveling direction detection sensor 30 is, for example, the road sign 4 shown in FIG. 2, and the object detected by the traveling direction detection sensor 30 includes a planar object such as the stop line 5 shown in FIG. It is. As the traveling direction detection sensor 30, for example, a front camera or a front radar can be used. The traveling direction detection sensor 30 also detects the shape of an object that exists in the traveling direction of the vehicle. The shape is detected by, for example, detecting distances to a plurality of points on the object.

  The traveling direction detection sensor 30 further outputs a traveling direction prediction error that is a value obtained by predicting the error of the traveling direction distance dy. Similar to the lateral direction detection sensor 20, the traveling direction prediction error is determined based on a signal used for detecting an object. The traveling direction prediction error also indicates a range including a measurement value of 95%, for example, and is specifically expressed as ± 10 cm. The traveling direction prediction error and the lateral direction prediction error determined by the lateral direction detection sensor 20 correspond to the detection error of the distance sensor in the claims.

  When the object detection range extends from the front of the vehicle to the side of the vehicle, an object detection sensor having both functions of the traveling direction detection sensor 30 and the lateral direction detection sensor 20 can be used.

  A high accuracy map 41 is stored in the storage unit 40. The high-precision map 41 is 3D map data in which the shape of a road and the shape of an object existing on and around the road are stored in association with absolute coordinates. Objects existing on the road are, for example, lane markings and road markings. Objects existing around the road are signs, buildings, and the like. Therefore, the objects detected by the lateral direction detection sensor 20 and the traveling direction detection sensor 30 are also objects whose coordinates are known in the high-precision map 41.

  The inertial navigation device 50 includes an inertial sensor such as a gyro sensor, and generates a travel locus determined based on the output value of the inertial sensor and the vehicle speed. Further, the inertial navigation device 50 corrects the position of the traveling locus based on the satellite positioning absolute position calculated by the GNSS receiver 10, thereby allowing the absolute position of the vehicle 2 (hereinafter, “inertial navigation absolute position”) and the traveling direction. Are sequentially determined. The inertial navigation absolute position and the heading can be sequentially determined even in a place where the navigation signal cannot be received.

  The inertial navigation apparatus 50 also determines an azimuth prediction error that means a predicted value of an error in the traveling azimuth. The azimuth prediction error is determined on the basis of the error of the inertial sensor and the elapsed time from the last correction of the position of the travel locus with the satellite positioning absolute position. For example, the azimuth prediction error is determined by multiplying the error per unit time of the inertial sensor by the elapsed time from the last correction of the position of the travel locus with the satellite positioning absolute position.

  The vehicle control unit 60 is a part that performs various controls on the vehicle. In this embodiment, the vehicle control unit 60 performs control based on the current position of the vehicle. The vehicle control unit 60 includes one or more vehicle control ECUs and includes a vehicle control application 61. The vehicle control application 61 is stored in a predetermined storage unit. The vehicle control application 61 is an application for executing various vehicle behavior controls based on the current position of the vehicle, and uses the corrected positioning absolute position output by the computing unit 70 as the current position. Further, the vehicle control application 61 determines whether or not to use the corrected positioning absolute position based on the positioning reliability output by the calculation unit 70.

  The calculation unit 70 is a computer including a CPU, a ROM, a RAM, and the like, and the CPU uses a temporary storage function of the RAM and stores it in a non-transitory tangible storage medium such as a ROM. By executing the stored program, the function as each unit shown in FIG. 1 is realized. That is, the calculation unit 70 functions as a matching unit 71, an antenna position offset determination unit 72, an antenna position prediction error calculation unit 73, a positioning position correction unit 74, and a positioning reliability output unit 75. When these functions are executed, a method corresponding to the program is executed. Note that part or all of the functions executed by the arithmetic unit 70 may be realized by hardware using one or a plurality of ICs.

  Next, the process of each part with which the calculating part 70 is provided is demonstrated. FIG. 3 shows processing executed by the matching unit 71. The execution timing of this process is, for example, every fixed period when the ignition is on.

  In step (hereinafter, step is omitted) S1, the shape of the landmark and the lateral distance dx of the landmark are acquired from the lateral detection sensor 20. The landmark means an object with which absolute coordinates are associated in the high-accuracy map 41, that is, an object whose absolute coordinates are known.

  In subsequent S2, the shape of the landmark and the traveling direction distance dy of the landmark are acquired from the traveling direction detection sensor 30. Note that the landmark for acquiring the traveling direction distance dy and the landmark for acquiring the lateral direction distance dx may be the same or different.

  In S3, the satellite positioning absolute position is acquired from the GNSS receiver 10. This processing corresponds to a positioning position acquisition unit in the claims. The lateral direction detection sensor 20 and the traveling direction detection sensor 30 cannot always detect a landmark. Therefore, the lateral distance dx and the traveling direction distance dy may not be acquired. Further, the GNSS receiver 10 may not be able to calculate the satellite positioning absolute position. Therefore, the satellite positioning absolute position may not be acquired.

  Therefore, in S4, S1 to S3 are executed to determine whether or not all of the lateral distance dx, the traveling direction distance dy, and the satellite positioning absolute position have been acquired. If this determination is NO, the process shown in FIG. 3 is terminated. On the other hand, if judgment of S4 is YES, it will progress to S5.

  In S5, the fixed range determined based on the satellite positioning absolute position acquired in S3 is set as a search range, and a high-accuracy map of the search range is acquired from the high-accuracy map 41.

  In S6, the absolute position of the vehicle 2 is determined by matching the high-accuracy map acquired in S5 with the landmark shape acquired in S1 and the landmark shape acquired in S2. The position of the vehicle 2 here means the position of the vehicle coordinate system origin O set in the vehicle 2 in advance. In the example of FIG. 2, the vehicle coordinate system origin O is the center point in the vehicle width direction on the front end surface of the vehicle 2. The vehicle coordinate system origin O corresponds to the reference position on the vehicle in the claims. Although not shown in FIG. 2, the height of the vehicle coordinate system origin O is also set in advance. The vehicle coordinate system origin O corresponds to a reference position in the claims. The vehicle coordinate system origin O may be another position of the vehicle 2.

  Hereinafter, the position of the vehicle 2 determined in S6 is referred to as a sensor positioning absolute position. The sensor positioning absolute position is represented by the latitude and longitude altitude. The x-coordinate of the sensor positioning absolute position is x1 + dx + w / 2 where w is the length in the width direction of the vehicle 2 and x1 is the x-coordinate of the landmark that has acquired the lateral distance dx. The y coordinate of the sensor positioning absolute position is y1-dy if the y coordinate of the landmark from which the traveling direction distance dy is acquired is y1. The z coordinate is also calculated in the same manner as the x and y coordinates. This S6 corresponds to a sensor positioning position determination unit in claims.

  FIG. 4 shows processing executed by the antenna position offset determination unit 72. The process of FIG. 4 is executed when the sensor positioning absolute position is determined in the process of FIG.

  In S11, the position difference is calculated by subtracting the sensor positioning absolute position determined in S6 from the satellite positioning absolute position acquired in S3. This position difference is the coordinate of the antenna 11 in a relative coordinate system centered on the vehicle coordinate system origin O.

  In S 12, which is a process corresponding to the azimuth obtaining unit in the claims, the traveling azimuth of the vehicle 2 is obtained from the inertial navigation device 50. Then, it progresses to S13. In addition, although illustration is abbreviate | omitted, when a vehicle azimuth | direction cannot be acquired, the process of FIG. 4 is complete | finished. Further, the process of FIG. 4 is not executed unless the sensor positioning absolute position is calculated in the process of FIG. Therefore, when the satellite positioning absolute position, the sensor positioning absolute position, and the vehicle direction cannot be acquired, the process does not proceed to S13.

  In S13, the position difference calculated in S11 is normalized. The normalization here is the coordinates of the antenna 11 in the vehicle coordinate system when the vehicle 2 is rotated so that the traveling direction of the vehicle 2 is north. Therefore, in S13, the position difference calculated in S11 is rotated by using the vehicle coordinate system origin O as the rotation center by the azimuth difference between the traveling direction of the vehicle 2 acquired in S12 and the north. The normalized coordinates are used as normalized antenna position offsets. This S13 corresponds to the antenna position determination unit in the claims, and north corresponds to the reference orientation in the claims. Of course, a direction other than the north may be used as the reference direction.

  The traveling direction of the vehicle 2 when the position difference is calculated may be various traveling directions. However, by normalizing the position difference and determining the normalized antenna position offset in this way, the antenna 11 can be located at any position on the vehicle 2 regardless of the traveling direction of the vehicle 2 when the position difference is calculated. Can be determined.

  In S14, it is determined whether or not the best antenna position prediction error has been updated. This best antenna position prediction error is updated in S29 of FIG. The best antenna position prediction error means the smallest antenna position prediction error after the ignition is turned on, among the antenna position prediction errors sequentially calculated by executing the processing of FIG.

  When the best antenna position prediction error is updated, it is considered that the antenna position prediction error is the smallest after the ignition is turned on, so that the antenna position offset is also considered to be the closest to the true value after the ignition is turned on. Therefore, if the determination in S14 is YES, the process proceeds to S15, and the best antenna position offset is updated to the normalized antenna position offset calculated in the immediately preceding S13. The best antenna position offset is stored in a predetermined offset storage unit such as a RAM.

  In the present embodiment, the best antenna position offset is reset when the ignition is turned on. This is because the antenna 11 may be moved while the ignition is turned off. Since the reset is performed when the ignition is turned on, the time when the ignition is turned on corresponds to the reference time point in the claims.

  FIG. 5 shows processing executed by the antenna position prediction error calculation unit 73. The process in FIG. 5 is executed when the sensor positioning absolute position is determined in the process in FIG. 3 as in the process in FIG.

  In S <b> 21, a lateral direction prediction error is acquired from the lateral direction detection sensor 20. In subsequent S <b> 22, a traveling direction prediction error is acquired from the traveling direction detection sensor 30. The lateral direction prediction error and the traveling direction prediction error are expressed as ± 10 cm as described above. In S23, a map error is acquired. The map error is a value representing an error of the high-precision map 41, and is a preset value. Map error is also expressed as ± 10 cm. The map error is stored in a storage unit inside the calculation unit 70 or a storage unit accessible by the calculation unit 70.

  In S24, a satellite positioning prediction error is acquired from the GNSS receiver 10. In S25, an azimuth prediction error is acquired from the inertial navigation apparatus 50. In S26, a normalization antenna position offset prediction error based on the azimuth prediction error (hereinafter, normalization offset prediction error) is calculated.

  The normalization offset prediction error will be described with reference to FIG. In FIG. 6, X is a distance from the vehicle coordinate system origin O to the antenna 11, and θ is an azimuth prediction error. Since there is an azimuth prediction error θ, even if the true position of the antenna 11 is P0, the position of the antenna 11 determined by the normalized antenna position offset calculated in S13 is some point from P1 to P2. There is a possibility. Since θ is a relatively small value, assuming that the arc P1-P0 and the arc P2-P0 are straight lines substantially orthogonal to the line segment O-P0, respectively, the distance from P1 to P0, and from P0 to P2 The distance is X · tan θ. Therefore, ± X · tan θ is a normalization offset prediction error.

  In S27, the antenna position prediction error is calculated by adding the five errors acquired or calculated in S21 to S24 and S26. The normalized antenna position offset calculated in S13 is calculated using two values of position difference and vehicle direction. Of these two values, the position difference is calculated from the lateral distance dx, the traveling direction distance dy, the high-precision map 41, and the satellite positioning absolute position. Therefore, these errors, that is, the lateral direction prediction error, the traveling direction prediction error, the map error, and the satellite positioning prediction error are added. Further, since the normalized antenna position offset is calculated using the vehicle azimuth, a normalization offset prediction error that is a prediction error of the normalized antenna position offset based on the azimuth prediction error is also added. The antenna position prediction error calculated in this way is a value obtained by predicting the normalized antenna position offset error calculated in S13.

  In S28, it is determined whether or not the antenna position prediction error calculated in S27 is smaller than the best antenna position prediction error stored in the prediction error storage unit. The prediction error storage unit may be the same storage unit as the above-described offset storage unit, or may be a different storage unit.

  If the determination in S28 is NO, the process of FIG. 5 is terminated as is, but if YES, the process proceeds to S29. In S29, the stored best antenna position prediction error is updated to the antenna position prediction error calculated this time. When the best antenna position prediction error is updated, the best antenna position offset is updated as described with reference to FIG. Note that this best antenna position prediction error is also reset when the ignition is turned on, as in the best antenna position offset.

  FIG. 7 shows processing executed by the positioning position correction unit 74. This process is executed every time the satellite positioning absolute position is acquired from the GNSS receiver 10 in S3. In S31, the best antenna position offset is acquired from the offset storage unit.

  In S32, the traveling direction of the vehicle 2 is acquired from the inertial navigation device 50. Alternatively, if FIG. 4 is executed subsequent to the processing of FIG. 3, the traveling direction of the vehicle 2 acquired in S <b> 12 may be used instead of acquiring the traveling direction of the vehicle 2 from the inertial navigation device 50. .

  Subsequent S33 and S34 correspond to a post-correction position determination unit. In S33, the antenna position offset at the time of correction is determined by rotationally correcting the best antenna position offset acquired in S31 based on the vehicle orientation acquired in S32.

  The best antenna position offset stored in the offset storage unit represents the antenna position in the vehicle coordinate system when the traveling direction of the vehicle 2 is facing north. Accordingly, the rotation correction here is a correction for rotating the best antenna position offset around the vehicle coordinate system origin O by the amount that the traveling direction of the vehicle 2 facing north becomes the direction acquired in S32.

  In S34, the corrected positioning absolute position is calculated by subtracting the corrected antenna position offset obtained in S33 from the acquired satellite positioning absolute position.

  FIG. 8 shows processing executed by the positioning reliability output unit 75. This process is also executed every time the satellite positioning absolute position is acquired from the GNSS receiver 10 as in FIG. 7 executed by the positioning position correction unit 74. Therefore, when the positioning position correction unit 74 calculates the corrected positioning absolute position, the positioning reliability output unit 75 executes the process of FIG.

  In S41, the best antenna position prediction error is acquired from the prediction error storage unit. The subsequent S42 corresponds to a satellite positioning prediction error acquisition unit, and acquires a satellite positioning prediction error from the GNSS receiver 10. Alternatively, if FIG. 5 is executed following the processing of FIG. 3, the satellite positioning prediction error acquired in S <b> 24 may be used instead of acquiring the satellite positioning prediction error from the GNSS receiver 10.

  In S43, the traveling direction prediction error is acquired from the inertial navigation apparatus 50. Alternatively, if FIG. 5 is executed subsequent to the processing of FIG. 3, the traveling direction prediction error acquired in S <b> 25 may be used instead of acquiring the traveling direction prediction error from the inertial navigation apparatus 50.

  In S44, an offset prediction error at the time of correction is calculated. Although the point at the time of correction is different from the offset prediction error at the time of normalization calculated at S26, the offset prediction error at the time of correction is calculated in the same manner as the offset prediction error at the time of normalization. That is, assuming that the bearing prediction error acquired in S43 is θ and the distance from the vehicle coordinate system origin O to the position of the antenna 11 determined by the best antenna position offset acquired in S31 is X, ± X · tan θ is This is an offset prediction error.

  In S45, the positioning reliability is calculated by adding the three errors acquired or calculated in S41, S42, and S44. S45 corresponds to a positioning reliability calculation unit in the claims. The positioning reliability calculated in this way is ± when the best antenna position prediction error acquired in S41 is e1, the satellite positioning prediction error acquired in S42 is e2, and the offset prediction error at the time of correction calculated in S44 is e3. (E1 + e2 + e3).

The corrected positioning absolute position calculated in S34 is calculated using the best antenna position offset, vehicle orientation, and satellite positioning absolute position. Positioning reliability calculated in S45, the addition of these best antenna position offsets, heading, satellite positioning absolute a corresponding error in the position best antenna position location prediction error, satellite positioning prediction error, the offset prediction error correction during doing. Therefore, the positioning reliability calculated in S45 represents the reliability of the corrected positioning absolute position.

  In subsequent S46, the corrected positioning absolute position calculated in S34 is output to the vehicle control unit 60 as the current position of the vehicle 2, and the positioning reliability calculated in S45 is also output to the vehicle control unit 60.

  In the present embodiment described above, the sensor positioning absolute position is sequentially determined in S6. This sensor positioning absolute position is the position of the vehicle 2 determined based on the distance between the vehicle 2 and the object whose absolute coordinates are stored in the high-precision map 41. Based on the position difference between this sensor positioning absolute position and the satellite positioning absolute position acquired in S3, a normalized antenna position offset representing the position of the antenna 11 with respect to the vehicle coordinate system origin O is calculated. Thus, in this embodiment, the position of the antenna 11 with respect to the vehicle coordinate system origin O can be calculated sequentially.

  In the positioning position correction process, the corrected positioning absolute position obtained by correcting the satellite positioning absolute position based on the antenna position offset is calculated as the position of the vehicle 2. Therefore, the position of the vehicle 2 that is not easily changed by the position of the antenna 11 can be output to the vehicle control unit 60.

  In addition, in this embodiment, the corrected positioning absolute value is based on the antenna position prediction error that is a predicted value of the antenna position offset error and the satellite positioning prediction error that is a predicted value of the satellite positioning absolute position error. The positioning reliability that is the reliability of the position is calculated. The calculation of the corrected positioning absolute position uses the antenna position offset and the satellite positioning absolute position, and therefore the positioning reliability calculated based on these two prediction errors, that is, the antenna position prediction error and the satellite positioning prediction error. Represents the reliability of the corrected positioning absolute position with high accuracy.

  In this embodiment, the traveling direction of the vehicle 2 is acquired in S12, and in S13, the normalized antenna position offset that is an antenna position offset when the traveling direction of the vehicle 2 is set to the north direction based on the traveling direction. Is calculated. When calculating the corrected positioning absolute position, the traveling azimuth of the vehicle 2 is acquired in S32, the best antenna position offset is rotationally corrected based on the traveling azimuth, and the corrected positioning absolute position is calculated. ing. Thereby, it is possible to calculate a corrected positioning absolute position with high accuracy regardless of the orientation of the vehicle 2 when calculating the corrected positioning absolute position.

  In the present embodiment, the positioning reliability that represents the reliability of the corrected positioning absolute position is not only the best antenna position prediction error and the satellite positioning prediction error, but also the prediction error of the antenna position offset caused by the azimuth prediction error. An offset prediction error at the time of correction is added. Thereby, it becomes the positioning reliability with high accuracy in consideration of the error of the vehicle direction.

  In the present embodiment, the antenna position prediction error is sequentially calculated, and the minimum antenna position prediction error after the ignition is turned on is set as the best antenna position prediction error. When this best antenna position prediction error is determined, it can be estimated that the antenna position offset error is the smallest. Therefore, the antenna position offset when determining the best antenna position prediction error is set as the best antenna position offset, and this best antenna position offset is used to calculate the corrected positioning absolute position. As a result, it is possible to calculate a corrected positioning absolute position with high accuracy.

  The antenna position prediction error used for calculating the positioning reliability is also set as the best antenna position prediction error. Further, an offset prediction error at the time of correction used for calculating the positioning reliability is also calculated using the best antenna position offset. Thereby, the accuracy of positioning reliability is further improved.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the above-mentioned embodiment, The following modification is also contained in the technical scope of this invention, Furthermore, the summary other than the following is also included. Various modifications can be made without departing from the scope. In the following description, elements having the same reference numerals as those used so far are the same as the elements having the same reference numerals in the previous embodiments unless otherwise specified. Further, when only a part of the configuration is described, the above-described embodiment can be applied to the other parts of the configuration.

<Modification 1>
In the above-described embodiment, the lateral direction prediction error and the traveling direction prediction error are dynamically changing values. However, one or both of the lateral direction prediction error and the traveling direction prediction error may be constant values.

<Modification 2>
In the above-described embodiment, the normalized antenna position offset is determined by normalizing the position difference. However, normalization is not essential if the position difference is stored as an antenna position offset as it is together with the vehicle orientation when the position difference is calculated.

<Modification 3>
The position of the vehicle coordinate system origin O need not be the center point in the vehicle width direction on the front end surface of the vehicle 2. For example, various positions such as the position of the center of gravity of the vehicle 2 and the mounting position of the antenna 11 at the time of shipment can be used as the vehicle coordinate system origin O.

<Modification 4>
In the above-described embodiment, the reference time point is set to the ignition-on time, but every fixed period may be set to the reference time point.

<Modification 5>
In the above-described embodiment, the positioning reliability is output to the vehicle control unit 60 together with the corrected positioning absolute position. However, only the corrected positioning absolute position may be output to the vehicle control unit 60.

1: Vehicle-mounted device 2: Vehicle 3: Road marking line 4: Road sign 5: Stop line 10: GNSS receiver 11: Antenna 20: Lateral direction detection sensor 30: Travel direction detection sensor 40: Storage unit 41: High-precision map 50 : Inertial navigation device 60: Vehicle control unit 61: Vehicle control application 70: Calculation unit 71: Matching unit 72: Antenna position offset determination unit 73: Antenna position prediction error calculation unit 74: Positioning position correction unit 75: Positioning reliability output unit

Claims (9)

A vehicle position determination device that is used in a vehicle equipped with a GNSS receiver that receives a navigation signal transmitted by a navigation satellite with an antenna, and determines the position of the vehicle,
A positioning position acquisition unit (S3) that sequentially calculates a satellite positioning absolute position calculated based on the navigation signal and that represents the position of the antenna in absolute coordinates;
A sensor positioning position determination unit (S6) for determining a sensor positioning absolute position that represents the position of the vehicle in absolute coordinates determined based on a distance between an object with known coordinates determined using a distance sensor and the vehicle;
An antenna position determination unit (S13) for determining an antenna position offset representing a position of the antenna with respect to a reference position on the vehicle based on a position difference between the satellite positioning absolute position and the sensor positioning absolute position;
A corrected position determining unit (S33, S34) for correcting the satellite positioning absolute position based on the antenna position offset and determining a corrected positioning absolute position;
An antenna position prediction error calculation unit that calculates an antenna position prediction error that is a value obtained by predicting an error of the antenna position offset determined by the antenna position determination unit, and determines the best antenna position prediction error based on the antenna position prediction error (73)
A satellite positioning prediction error acquisition unit (S42) that acquires a satellite positioning prediction error that is a value predicted from the error of the satellite positioning absolute position;
A vehicle position determination apparatus comprising: a positioning reliability calculation unit (S45) that calculates a positioning reliability that is a reliability of the corrected positioning absolute position based on the best antenna position prediction error and the satellite positioning prediction error. In claim 1,
An azimuth acquisition unit (S12, S32) that sequentially acquires the traveling azimuth of the vehicle,
The antenna position determination unit calculates a normalized antenna position offset that is the antenna position offset when the vehicle faces a reference direction based on the traveling direction when the position difference is calculated;
The post-correction position determination unit determines the antenna position offset at the time of correction based on the normalized antenna position offset determined by the antenna position determination unit and the traveling direction acquired by the direction acquisition unit at the time of correction, A vehicle position determination device that determines the corrected positioning absolute position from the determined antenna position offset at the time of correction and the satellite positioning absolute position acquired by the positioning position acquisition unit at the time of correction. In claim 2 ,
The positioning reliability calculation unit calculates the positioning reliability using a prediction error of a traveling direction of the vehicle at the time of calculating the positioning reliability in addition to the best antenna position prediction error and the satellite positioning prediction error. A vehicle position determination device. In any one of claims 1 to 3,
The antenna position prediction error calculation section, the antenna position sequentially calculates the prediction error, the vehicle position determination apparatus shall be the smallest of the antenna position prediction error the best antenna position prediction error in the subsequent predetermined reference time. In claim 4 ,
The antenna position determination unit, the antenna position offset when the antenna position prediction error calculation unit has determined the best antenna position prediction error, the best antenna position offset,
The post-correction position determination unit corrects the satellite positioning absolute position acquired by the positioning position acquisition unit based on the best antenna position offset to determine the corrected positioning absolute position. In any one of Claims 1-5 ,
The distance sensor is mounted on the vehicle and detects the shape of the object outside the vehicle in addition to detecting the distance to the object outside the vehicle,
The sensor positioning position determination unit collates the shape of the object outside the vehicle detected by the distance sensor with map data in which the shape of an object existing on and around the road is stored in association with coordinates. The vehicle position determination apparatus which determines the said sensor positioning absolute position by this. In any one of Claims 1-6 ,
The antenna position prediction error calculating unit calculates the antenna position prediction error based on a detection error of the distance sensor and a prediction error of the satellite positioning absolute position. In claim 2 or 3 ,
The antenna position prediction error calculation unit calculates the antenna position prediction error based on the detection error of the distance sensor, the prediction error of the satellite positioning absolute position, and the prediction error of the traveling direction of the vehicle. Decision device. In claim 6 ,
The antenna position prediction error calculation unit calculates the antenna position prediction error based on a detection error of the distance sensor, a prediction error of the satellite positioning absolute position, and an error of the map data.

Images