JP7080283B2 - Target route generator and target route generation system - Google Patents

Description

The present application relates to a target route generation device and a target route generation system.

Technologies that can estimate accurate vehicle position information using information from in-vehicle devices such as GNSS (Global Navigation Satellite System) receivers, cameras, and millimeter-wave radar are becoming widespread, and high-precision maps including lane-level information are available. By combining it and utilizing it in an automatic driving system, it is expected to realize highly accurate automatic driving.
On the other hand, the high-precision map compatible with the automatic driving system has a limited area such as a highway, and the range of utilization of automatic driving using the high-precision map is limited.
Therefore, there has been proposed a method of recording travel locus information by estimating accurate position information when the own vehicle is manually driven, and converting the recorded travel locus information into high-precision map information.
For example, in Patent Document 1, a plurality of track information is collected by traveling of a vehicle, a road having a plurality of collected track information is divided into a plurality of parts, and characteristic points are extracted from each of the divided road parts. A method of generating a high-precision map as the center line of a lane by calculating a spline curve passing through the extracted feature points is described.

Japanese Patent No. 5064870 (pages 7 to 12, FIG. 3)

The vehicle position information acquired by the in-vehicle device varies due to measurement errors of the in-vehicle device and the like. In Patent Document 1, a method is adopted in which an error between a spline curve passing through a feature point and an actual center line of a lane is converged by collecting a plurality of locus information.
However, in order to calculate a spline curve with a small error from the actual center line of the lane, it is necessary to collect a lot of track data, and if the number of track data to be collected is small, the calculated vehicle center line is calculated. There is a problem that the curve becomes a distorted shape.

The present application discloses a technique for solving the above-mentioned problems, and is a target route generation device and a target route generation system capable of generating a target route having a smooth curve regardless of the number of times of data collection of the vehicle travel locus. The purpose is to provide.

The target route generation device disclosed in the present application is a vehicle travel information acquisition unit that acquires vehicle travel information including vehicle position information when the vehicle is manually driven, and a vehicle travel information acquired by this vehicle travel information acquisition unit. , A vehicle travel locus generator that generates a vehicle travel locus in which each position information is arranged in chronological order, and a vehicle that divides the vehicle travel locus generated by this vehicle travel locus generation unit into a plurality of sections based on the position information. Based on the travel locus division unit and the vehicle travel locus divided by this vehicle travel locus division unit, the approximate curve of the travel locus when the vehicle is manually driven is to be continuous for each section and at the boundary of the adjacent section. , The approximate curve calculation unit to be calculated with the end point of each section as the start point of the next section, and the target route to generate the target route for vehicle traveling by combining the approximate curves of each section calculated by this approximate curve calculation unit. It is equipped with a generator.

According to the target route generation device disclosed in the present application, it is possible to generate a target route having a smooth curve regardless of the number of times of data acquisition of the vehicle travel locus.

It is a block diagram which shows the structure of the target route generation apparatus by Embodiment 1. FIG. It is a flowchart which shows the whole operation of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the example of the selection processing of the vehicle running information of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the example of the selection processing result of the vehicle running information of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the division | division section of the vehicle travel locus of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the coordinate rotation in the 1st section of the target path generation apparatus by Embodiment 1. FIG. It is a flowchart which shows the process of determining the section length of the Nth section of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the method of determining the section length of the Nth section by the directional difference information of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the calculation of the approximate curve in the 1st section of the target route generation apparatus by Embodiment 1. FIG. It is a figure which shows the calculation of the approximate curve after the 2nd section of the target route generation apparatus by Embodiment 1. FIG. It is a block diagram which shows the structure of the target route generation apparatus by Embodiment 2. FIG. It is a flowchart which shows the whole operation of the target route generation apparatus by Embodiment 2. FIG. It is a figure which shows the calculation of the approximate curve using the vehicle running information of the target route generation apparatus and the target route by Embodiment 2. FIG. It is a block diagram which shows the structure of the target route generation system by Embodiment 3. FIG. FIG. 3 is a block diagram showing another configuration of the target route generation system according to the third embodiment. It is a figure which shows the hardware composition of the target route generation apparatus by Embodiment 1 and Embodiment 2.

Embodiment 1.
Hereinafter, the first embodiment will be described with reference to the drawings. In the following, the vehicle center line will be referred to as the target route.
FIG. 1 is a block diagram showing a configuration of a target route generation device according to the first embodiment.
In FIG. 1, the target route generation device 100 is mounted on a vehicle and is configured as follows. A vehicle equipped with the target route generation device 100 is referred to as a "own vehicle".
The vehicle travel information acquisition unit 11 acquires vehicle travel information including position information when the own vehicle is manually traveled. The position information is acquired by receiving the positioning signal by the satellite positioning system at a predetermined time interval.

The vehicle travel locus generation unit 12 creates a vehicle travel locus of its own vehicle by using the vehicle travel information acquired by the vehicle travel information acquisition unit 11. Here, the vehicle travel locus is a arrangement of driving information (vehicle travel information) acquired by the vehicle travel information acquisition unit 11 in chronological order.
The vehicle travel locus dividing unit 13 divides the vehicle travel locus generated by the vehicle travel locus generation unit 12 into a plurality of sections.
The approximate curve calculation unit 14 calculates an approximate curve of an actual travel locus by using the information of the vehicle travel locus in each section divided by the vehicle travel locus dividing unit 13.
The target route generation unit 15 generates a target route for vehicle traveling by combining the approximate curves which are the approximate curve calculation results of each section calculated by the approximate curve calculation unit 14.

FIG. 3 is a diagram showing an example of selection processing of vehicle travel information of the target route generation device according to the first embodiment.
In FIG. 3, the position information displayed as dots in the figure constituting the vehicle traveling information is information indicating the position of the own vehicle measured by the satellite positioning system. Longitude information is represented as Lon _p , and latitude information is represented as Lat _p . The actual travel locus 21 is a locus in which the own vehicle actually traveled.
FIG. 3A is a diagram showing the position information of the own vehicle whose longitude information is Lon _p and whose latitude information is Lat _p , which is acquired at regular time intervals. At this time, the position information acquired due to the measurement error of the satellite positioning system does not always match the actual traveling locus 21 of the own vehicle.
FIG. 3B excludes this position information so as not to be reflected in the generation process of the target route for vehicle travel when it is estimated that the position information far away from the actual travel locus 21 is acquired. It is a figure which shows the generated vehicle travel locus. In FIG. 3B, the position information of the longitude information Lon ₄ , the latitude information Lat ₄ , and the longitude information Lon ₇ and the latitude information Lat ₇ in FIG. 3A are excluded. The vehicle travel locus in FIG. 3B is a time-series arrangement of vehicle travel information, and is represented as n ₁ , n ₂ , n ₃ ... n _p in order from the earliest acquired vehicle. ..

FIG. 4 is a diagram showing an example of a selection processing result of vehicle travel information of the target route generation device according to the first embodiment.
FIG. 4 is an example in which the reliability corresponding to the position information acquired by the satellite positioning shown in the example of FIG. 3 is also shown. In FIG. 4, the reliability of the longitude information Lon ₄ , the latitude information Lat ₄ , the longitude information Lon ₇ , and the latitude information Lat ₇ is the lowest determination, and it is shown that these are excluded.

FIG. 5 is a diagram showing a divided section of the vehicle travel locus of the target route generation device according to the first embodiment.
In FIG. 5, reference numeral 21 is the same as that in FIG.
FIG. 5A is a diagram showing division of the vehicle traveling locus of the own vehicle shown in FIG. The vehicle traveling loci arranged in chronological order are divided and represented as the first section, the second section, the third section, and so on in order from the earliest.
FIG. 5B is a diagram showing division in the case of having a region 31 in which the acquisition interval of vehicle traveling information is wide open. If there is a region 31 in which the acquisition interval of vehicle travel information is wide, the section of the vehicle travel locus having this region 31 is excluded from the target route generation target.

FIG. 6 is a diagram showing coordinate rotation in the first section of the target route generation device according to the first embodiment.
FIG. 6A is a diagram in which longitude information and latitude information are converted using an ENU (East North Up) coordinate system having an X-axis in the east direction and a Y-axis in the north direction. At this time, the origin of the ENU coordinate system is set so as to coincide with the position of the vehicle traveling information with the earliest acquisition timing.
FIG. 6B is a diagram showing that the coordinates are rotated with respect to the ENU coordinate system of FIG. 6A to determine the coordinate system of the first section. This rotation process of the coordinate axes is performed in order to arrange the vehicle traveling locus so that the calculation of the approximate curve can be surely performed.

FIG. 8 is a diagram showing a method of determining the section length of the Nth section based on the directional difference information of the target route generation device according to the first embodiment.
FIG. 8A is a diagram showing a method of deriving the directional difference Δθ _p . In order to derive the directional difference Δθ _p , the directional information θ _p with respect to the vehicle traveling information n _p is derived. The azimuth information θ _p consists of a straight line connecting the currently extracted vehicle travel information n _p and the vehicle travel information n _{p + 1} one ahead of it, and Y ₁ which is the Y axis of the coordinate system in the first section. Represents an angle.
The directional difference Δθ _p is the difference in angle between the directional information θ p of the currently extracted vehicle running information and the directional information θ _{p -1} of the vehicle running information immediately before it | (θ _p )-(θ _p ). _-1 ) | The value of Δθ ₁ is 0.
FIG. 8B is a diagram showing a method of deriving the first section length. The start point is the position where the vehicle travel information n ₁ exists, the end point is the position where the vehicle travel information n _p-1 exists, and the range is parallel to the X axis (X ₁ ) in the coordinate system of the first section.

FIG. 9 is a diagram showing the calculation of an approximate curve in the first section of the target route generation device according to the first embodiment.
FIG. 9 shows an image in which the derived approximate curve and the vehicle traveling locus are superimposed. The end point of the first section length set in the stage before the approximate curve calculation process is the position coordinates of the vehicle traveling information n _p-1 . At the stage after the calculation of the approximate curve, the end point of the first interval is a point (x _1e , y _1e ) existing on the approximate curve 22.

FIG. 10 is a diagram showing the calculation of the approximate curve after the second section of the target route generation device according to the first embodiment.
In FIG. 10, the approximate curve 22a is the approximate curve of the first section, the approximate curve 22b is the approximate curve of the second section, and the approximate curve 22c is the approximate curve of the third section.
The coordinate system of the second section is translated with respect to the ENU coordinate system so that the end point (x _1e , y _1e ) of the first section derived earlier coincides with the origin. Further, in the coordinate rotation, the value of the first derivative at the end point of the approximate curve in the first section is rotated so as to be parallel to the X axis (X ₂ ) of the coordinate system in the second section.

Next, the operation will be described.
The overall operation of the target route generation device according to the first embodiment will be described with reference to the flowchart of FIG.
First, in step S101, the vehicle travel information acquisition unit 11 acquires vehicle travel information. The vehicle travel information is acquired by the user manually driving the own vehicle equipped with the target route generation device 100. When the own vehicle is driven by the user's manual operation, as shown in FIG. 3A, the position information of the own vehicle whose longitude information is Lon _p and whose latitude information is Lat _p is acquired at regular time intervals. At this time, the position information acquired due to the measurement error of the satellite positioning system does not always match the actual traveling locus of the own vehicle.

In step S102, sorting processing is performed on the vehicle traveling information acquired in step S101. This sorting process is executed by the vehicle travel locus generation unit 12. The received information by satellite positioning includes satellite positioning mode information in addition to latitude information and longitude information.
By utilizing this satellite positioning mode information, it is possible to determine the reliability of the received latitude information and longitude information.
FIG. 4 is an example in which the reliability corresponding to the position information acquired by the satellite positioning shown in FIG. 3 is also shown. In the example of FIG. 4, the reliability in the longitude information Lon ₄ , the latitude information Lat ₄ , the longitude information Lon ₇ , and the latitude information Lat ₇ is the lowest determination.
This is because it is presumed that the position information far away from the actual travel locus 21 is acquired as in the example of FIG. 3A, and this position information is not reflected in the target route generation process. I try to exclude it. As a result, the target route generated by the approximate curve calculation process can have a smaller error in position with respect to the actual vehicle travel locus.

Next, in step S103, a process of generating a vehicle traveling locus is performed.
Here, the vehicle travel locus with highly reliable position information left by the sorting process performed in step S102 is generated according to the measured time series.
As shown in FIG. 3B, the vehicle travel information included in the vehicle travel locus is represented as n ₁ , n ₂ , n ₃ ... n _p in order from the earliest acquired vehicle in chronological order. To.

Subsequent processes from step S104 to step S106 are executed by the vehicle travel locus dividing unit 13.
In step S104, the process of determining the first section is performed with N = 1. As shown in FIG. 5A, the vehicle travel loci divided by the vehicle travel locus dividing unit 13 are represented as a first section, a second section, and a third section in order from the earliest in chronological order. ..
The process of determining the section length of each section is performed in the order of the first section, the second section, the third section, and the Nth section. Therefore, in step S104, first, N = 1 is set, and a process of determining the first section is performed.

Next, in step S105, the process of determining the coordinate system in the first section is performed. In order to derive an approximate curve from the longitude information and latitude information acquired by satellite positioning, it is necessary to represent the longitude information and latitude information in a Cartesian coordinate system.
Here, as shown in FIG. 6A, longitude information and latitude information are converted using an ENU coordinate system having an X-axis in the east direction and a Y-axis in the north direction. At this time, the origin of the ENU coordinate system is set so as to coincide with the position of the vehicle traveling information with the earliest acquisition timing.

After determining the ENU coordinate system, as shown in FIG. 6B, the coordinates are rotated with respect to the determined ENU coordinate system to determine the coordinate system of the first section. The rotation process of the coordinate axes is performed in order to arrange the vehicle traveling locus so that the approximate curve can be calculated reliably.
The approximation curve calculation unit 14 performs a process of approximating with a cubic function represented by y = A × x ³ + B × x ² + C × x + D. However, for example, when the vehicle travel locus is heading straight north, that is, when the vehicle travel locus is parallel to the Y axis, the coefficients A, B, C represented by the cubic function, D cannot be derived.
Therefore, by rotating the coordinates so that the point cloud becomes a traveling locus that is almost parallel to the X axis, it is possible to derive the coefficient of the cubic function for the locus in all traveling directions. I am doing it.

For this reason, the process of deriving the rotation angle of the coordinate system is performed.
As a derivation method, as shown in FIG. 6 (b), a straight line connecting the origin, that is, the vehicle traveling information with the earliest acquisition timing, and the vehicle traveling information other than the origin (in the example of FIG. 6 (b), (x ₄ , y). ₄ )) is connected to), but there is a method of rotating the coordinate system so that it is parallel to the X axis (X1 axis) in the _first section.
However, this is only an example, and as another method, a plurality of angles formed by a straight line connecting two different vehicle traveling information and the X-axis may be derived, and the average value thereof may be determined as the rotation angle.

Next, in step S106, a process of determining the length (section length) of the first section is performed.
The details of the process of step S106 will be described later using the flowchart of FIG. 7.

Next, in step S107, an approximate curve of the first section having the section length derived by step S106 is derived. The approximate curve to be derived is a cubic function shown in Eq. (1).
y ₁ (x) = C0 ₁ + C1 ₁ x + C2 ₁ x ² + C3 ₁ x ³ ... (1)

y1 (x) represents the approximate curve 22a in the first interval, and C0 ₁ , C1 ₁ , C2 ₁ , and C3 ₁ represent the coefficients of the cubic function. C0 ₁ , C1 ₁ , C2 ₁ , and C3 ₁ are derived by solving the least squares method for the vehicle traveling information existing in the first section.
In this derivation method, the evaluation function of the equation (2) is used to obtain the coefficients C0 ₁ , C1 ₁ , C2 ₁ , and C3 ₁ such that the value _feva of the evaluation function is minimized. However, (x _1i , y _1i ) represents each position coordinate of the vehicle traveling locus in the first section.

FIG. 9 shows an image in which the derived approximate curve and the vehicle traveling locus are superimposed.
The end point of the first section length set in the stage before the approximate curve calculation process is the position coordinates of the vehicle traveling information n _p-1 , but in the stage after the approximate curve calculation, the end point of the first section is the approximate curve. It is a point (x _1e , y _1e ) existing on 22.

After the calculation process of the approximate curve is completed, the process proceeds to step S108, and a process of adding weight information to the generated approximate curve is performed.
In the first embodiment, the weight information to be given is 1. The handling of weight information will be described later in the second embodiment.

After the weight information assignment process, the process proceeds to step S109. In step S109, it is determined whether or not all the acquired vehicle travel information is used for the calculation of the approximate curve. For example, in the case of FIG. 9, unused vehicle travel information exists after the first section. In this case, the process proceeds to step S110, N = 2, returns to step S105, and processing for the second section. I do.

The processing method for the second and subsequent sections is basically the same as the processing from step S105 to step S108 described above.
However, regarding the process of determining the coordinate system in step S105, only the coordinate rotation process was performed on the ENU coordinate system in the first section, but the coordinate rotation process was performed on the ENU coordinate system in the second and subsequent sections. In addition to that, the coordinate translation process is also performed.

Next, the process of determining the coordinate system (step S105) in the second and subsequent sections will be described with reference to FIG.
The coordinate system of the second section is translated with respect to the ENU coordinate system so that the end point (x _1e , y _1e ) of the first section derived earlier coincides with the origin.
Regarding the coordinate rotation, the value of the first derivative at the end point of the approximate curve 22a in the first section is rotated so as to be parallel to the X axis (X2) of the coordinate system in the _second section.
As a result, the approximate curve in the second interval is expressed in the form of Eq. (3) in which N = 2 and 0 is substituted for the coefficients C0 and C1 with respect to Eq. (1).
Further, the relationship between the equation (4) and the equation (5) is established at the boundary between the approximate curve of the first section and the approximate curve 22b of the second section.
y ₂ (x) = C2 ₂ x ² + C3 ₂ x ³ ... (3)
y ₁ (x _1e ) = y ₂ (0) ・ ・ ・ ・ ・ (4)

By using the relational expressions of equations (4) and (5), the approximate curves of both sections are continuously connected, and the approximate curves are interrupted or discontinuously connected at the boundary of the sections. It is possible to avoid such things.

In the third and subsequent sections, the same relational expressions as those in the equations (3), (4), and (5) are used. The approximate curve in the Nth interval is expressed by the equation (6), and the relational expression at the origin of the Nth interval is expressed by the equations (7) and (8).
y _N (x) = C2 _N x ² + C3 _N x ³ ... (6)
y _N-1 (x _N-1e ) = y _N (0) ・ ・ ・ ・ ・ (7)

The evaluation function in the Nth interval is given by Eq. (9). However, (x _Ni , y _Ni ) represents each position coordinate of the vehicle traveling locus in the Nth section.

The processes from step S105 to step S108 are repeated in the second and subsequent sections as well.
When all the acquired vehicle travel information is used in the calculation of the approximate curve in step S109, the process proceeds to step S111, and the target route generation unit 15 performs the target route generation process.
Here, the process of combining the approximate curves derived in each section is performed, and the route is generated as one target path.
However, when the end point of the section is set from the distance determination of the vehicle travel information in the process of step S204 described later, as shown in FIG. 5 (b), the section between the first section and the second section is combined. Do not do this, and set it as a section that is not subject to target route generation.

Next, the details of the processing contents of step S106 executed by the vehicle traveling locus dividing unit 13 will be described with reference to the flowchart shown in FIG. 7.
FIG. 7 shows a process of determining the section length of the Nth section of the target route generator.
First, in step S201, a portion to be the starting point of the section is determined. The starting point of the _first section is the position of the vehicle traveling information n1.

Next, the process of determining whether or not the vehicle traveling information after n ₂ is within the first section is performed from step S202 to step S208.
In steps S202 and S203, vehicle traveling information n ₂ , n ₃ , n ₄ ... Is extracted in this order.

Next, in step S204, the distance D _p when the currently extracted vehicle travel information n _p and the previous vehicle travel information n _p-1 are connected by a straight line is derived, and D _p is derived. It is determined whether or not it is equal to or less than the preset upper limit value Dmax.

If the distance D _p is larger than the upper limit value D max in step S204, the process proceeds to step S209, and the vehicle travel information n _p-1 immediately before the currently extracted vehicle travel information is set to the first section. Let it be the end point of.
The reason for performing the distance determination process of the vehicle travel information is to prevent the calculation result of the approximate curve from being significantly different from the actual travel locus 21.
As shown in FIG. 5A, in the case of a vehicle travel locus in which the acquisition intervals of vehicle travel information are dense, it is expected that an approximate curve with a small error can be obtained with respect to the actual travel locus 21. ..
On the other hand, as shown in FIG. 5B, in the region 31 where the acquisition interval of the vehicle travel information is wide, when the approximate curve calculation is performed using the vehicle travel information existing before and after this region 31, it is actually performed. There is a possibility that the result of the approximate curve calculation will be significantly different from the traveling locus of.
Therefore, as shown in FIG. 5B, when a region 31 in which the acquisition interval of vehicle traveling information is wide is detected, immediately before this section is set as the end point of the Nth section, and immediately after this section is the N + 1th section. Performs processing to set the starting point of.

If it is determined in step S204 that the distance D _p is equal to or less than the upper limit value D max, the process proceeds to step S205.

When deriving an approximate curve from a vehicle traveling locus, there is a characteristic that the error due to the cubic function approximation increases as the curvature increases. In a vehicle traveling locus with a large curvature, it is desirable to shorten the approximation section in order to reduce the approximation error.
Steps S205 to S207 are processes for this purpose.

In step S205, a process of deriving the directional difference Δθ _p between the currently extracted vehicle travel information n _p and the previous vehicle travel information n _p-1 is performed.
The method of deriving the directional difference Δθ _p is shown in FIG.
As shown in FIG. 8A, in order to derive Δθ _p , the directional information θ _p with respect to the vehicle traveling information n _p is derived. The azimuth information θ _p consists of a straight line connecting the currently extracted vehicle travel information n _p and the vehicle travel information n _{p + 1} one ahead of it, and Y ₁ which is the Y axis of the coordinate system in the first section. Represents an angle.
Then, the difference in angle between the directional information θ p of the vehicle running information currently extracted and the directional information θ _{p -1} of the vehicle running information immediately before it | (θ _p )-(θ _p-1 ) | Let the value represented by be the azimuth difference Δθ _p . The value of Δθ ₁ is 0.

In the following step S206, the cumulative value of the directional difference (( _{Δθ 1} ) + (Δθ ₂ ) + (Δθ ₃ ) + ...・) Is derived.

Next, in step S207, the cumulative value of the directional difference derived in step S206 is compared with the preset angle threshold value (Δθ _th ).
If the cumulative value (ΣΔθ _p ) of the directional difference up to the currently extracted vehicle traveling information _np is equal to or less than the angle threshold value (Δθ _th ), the process proceeds to step S208.

In step S208, it is determined that the vehicle traveling information _np exists in the first section.
Then, the process returns to step S202, the extracted vehicle travel information is changed from n _p to n _{p + 1} , and the processes after step S203 are repeated.

If the cumulative value (ΣΔθ _p ) of the directional difference up to the currently extracted vehicle travel information _np in step S207 is larger than the angle threshold value (Δθ _th ), the vehicle travel information _np is outside the first section. It is determined that the vehicle exists in, and the process proceeds to step S209, and n _p-1 , which is one before the currently extracted vehicle travel information, is set as the end point of the first section.

The reason why the cumulative value of the directional difference is used in the determination of the end point of the section in step S207 is to set the range so that an approximate curve having a small error with respect to the actual traveling locus can be derived.
For example, if the actual traveling locus of the vehicle is shaped like a fan-shaped arc with a central angle of about 90 degrees, it is not possible to perform an approximate curve calculation with a small error using the cubic function y = Ax ³ + Bx ² + Cx + D. It is difficult due to the characteristics of the function. In order to perform an approximate curve calculation with a small error using a cubic function, in the case of an arc, it is necessary to divide the arc so that the central angle becomes small and perform an approximate curve calculation for each arc.
In this way, by limiting the cumulative value of the directional difference, it is possible to divide the traveling locus, which is difficult to approximate as described above, into an appropriate length.

After the end point of the Nth section is derived in step S209, the section length is derived in step S210.
As for the method of deriving the first section length, as shown in FIG. 8B, the first point is the position where the vehicle travel information n ₁ exists and the end point is the position where the vehicle travel information n _p-1 exists. The range is parallel to the X axis (X ₁ ) in the coordinate system of the interval.
The above is the process of determining the section length of the Nth section shown in FIG.

In the above description, the cumulative value of the directional difference information of the vehicle traveling information is used as the method of determining the Nth section length, but a method of using the vehicle speed information instead of using the directional difference information may be used.
In a traveling locus with a large curvature that makes it difficult to approximate a cubic function, the vehicle basically travels at a low vehicle speed, so the section length according to the vehicle speed can be shortened at low speeds by utilizing this. May be set.

Further, in the above description, the coordinate position and the value of the first derivative are equal at the end point of a certain section and the start point of the next section so that the approximate curves of adjacent sections are continuously connected. However, in order to connect more smoothly, the approximate curve may be derived by adding the condition that the values of the second derivative are equal.

According to the first embodiment, smooth curve data is derived in each divided section, and approximate curves are continuously connected at the boundary of the section, so that the generated target path is a smooth curve. It becomes.
Therefore, regardless of the number of times the vehicle travel information is acquired, a target route having a smooth curve can always be obtained, and an appropriate target route for vehicle travel can be generated.

Embodiment 2.
Next, the second embodiment will be described with reference to the drawings.
FIG. 11 is a block diagram showing the configuration of the target route generation device according to the second embodiment.
In FIG. 11, reference numerals 11 to 15 and 100 are the same as those in FIG. In FIG. 11, the target route generated by the target route generation unit 15 is input to the approximate curve calculation unit 14.
As a result, both the target route output from the target route generation unit 15 and the divided vehicle travel locus output from the vehicle travel locus division unit 13 are input to the approximate curve calculation unit 14.

FIG. 13 is a diagram showing the calculation of an approximate curve using the vehicle travel information of the target route generation device according to the second embodiment and the target route.
In FIG. 13, the information represented by the dots is the vehicle travel locus generated by the vehicle travel locus generation unit 12. The target route 23 is an existing target route generated in the past, and is drawn in the vicinity of the vehicle traveling locus. The target route extraction area 32 is an area within the range of the first section, and the target route in this area is extracted.
FIG. 13A is a diagram showing the target route 23 at the stage where the first section length is determined.
FIG. 13B is a diagram in which a plurality of points existing on the function of the target path 23 are extracted. The score extracted at this time is the same as the score of the vehicle traveling information existing in the Nth section.
FIG. 13C is a diagram showing an approximate curve 22 of the Nth section derived using both the vehicle travel locus and the target route. The approximation curve 22 is closer to the target path 23 due to the weight information given to the target path described later.

Next, the operation will be described.
The overall operation of the target route generation device according to the second embodiment will be described with reference to the flowchart of FIG.
The processing of steps S301 to S306 is the same as the processing of steps S101 to S106 of FIG. 2, and the description thereof will be omitted.

In step S307, a process of determining whether or not a target route exists within the range of the Nth section derived in step S306 is performed.
FIG. 13A shows a stage in which the first section length is determined by step S306, and the vehicle travel locus generated by the vehicle travel locus generation unit 12 is represented by dots. The curve drawn near the vehicle travel locus is the target route 23 output from the target route generation unit 15. The target route 23 is generated at a past timing different from the acquisition timing of the vehicle traveling locus in the figure.
Therefore, while the vehicle travel locus is new data, the target route 23 can be paraphrased as existing data.

As _a method of determining whether or not the target route exists within the range of the Nth section, as shown in FIG. 13A, the direction of the X1 axis is the section length x _1a and the _Y1 axis with respect to the section coordinates. The direction of is determined by drawing a target route extraction area 32 having the set value y _1a as an end and determining whether or not the target route is included in the target route extraction area 32.
The value of y _1a may be a preset fixed value or may be variable according to the section length x _1a . Further, it may be variable according to the number of vehicle traveling information included in the section.

If it is determined in step S307 that the target route exists within the range of the Nth section, the process proceeds to the process of step S308, and the vehicle position information and the weight information are obtained from the target route 23 existing within the range of the Nth section. To extract.
Here, as for the vehicle position information, as shown in FIG. 13B, a plurality of points existing on the function of the target route 23 are extracted. The score extracted at this time is the same as the score of the vehicle traveling information existing in the Nth section. For the weight information described later, the value given to the target route is extracted.

After that, the process proceeds to step S309, and as shown in FIG. 13 (c), the approximate curve 22 of the Nth section is calculated using both the vehicle travel locus and the target route 23.
The calculated approximate curve 22 is a cubic function as in Eqs. (1) and (6), and each coefficient is derived by solving the least squares method. However, unlike the case of the first embodiment, the data used when solving the least squares method is both the vehicle travel locus and the target route 23.

The evaluation function of the least squares method using both the vehicle travel locus and the target route 23 is expressed by Eq. (10). In equation (10), (x _Ni , y _Ni ) is the position coordinates of the vehicle traveling locus in the Nth section, and (x _Nj , y _Nj ) is each of the information extracted from the target route in the Nth section. Represents position coordinates.
As in the case of the first embodiment, a coefficient is derived so that the value _feva of the evaluation function is minimized.

Here, W in the equation (10) represents weight information. The weight information W is applied only to the coordinates of each position information extracted from the target path. In step S108 of the first embodiment, the weight information 1 is given, but in this case, there is no difference in the coefficients for the 1st item and the 2nd item of the evaluation function and the equation (10), and neither the vehicle traveling locus nor the target route is obtained. The least squares method with the same weight is performed.
On the other hand, when the weight information W has a value larger than 1, in order to derive the minimum value by the evaluation function, a position closer to the position coordinates extracted from the target route than the position coordinates of the vehicle traveling locus is selected. Coefficients will be derived so that an approximate curve to pass through is drawn.
For example, FIG. 13C shows a case where the weight information 2 is added to the target route 23, and the derived approximate curve 22 is closer to the target route 23.
That is, by setting the weight information applied to the target route to a value larger than 1, it is possible to adjust how strongly the information of the target route is reflected in the calculation result of the approximate curve.

In step S310, a process of adding weight information to the approximate curve derived in step S309 is performed.
At this time, a value different from the weight information given to the target path before the approximate curve calculation can be given. For example, the approximate curve calculation process is performed using the target path before the approximate curve calculation to which the weight information 1 is given, and the weight information 2 is given to the obtained approximate curve as a result.
By increasing the value of the weight information in proportion to the number of times the approximate curve calculation is performed in this way, the content of the target path is more strongly reflected in the subsequent approximate curve calculation processing.

On the other hand, if it is determined in step S307 that the target route does not exist within the range of the Nth section, the process proceeds to steps S311 and S312.
The processing contents of steps S311 and S312 are the same as the processing of steps S107 and S108 in the first embodiment.

Next, in step S313, it is determined whether or not all the vehicle travel information has been used in the calculation of the approximate curve, and if there is unused vehicle travel information in the calculation of the approximate curve, N = N + 1 in step S314. The process from step S305 to step S310 is repeated.

When all the vehicle traveling information is used for the calculation of the approximate curve, the process proceeds to step S315, and the acquired approximate curves of each section are combined to generate a target route.

According to the second embodiment, when the target route has already been acquired, the vehicle travel locus is newly acquired to reflect the newly acquired vehicle travel locus information for the target route, which is an updated version. A target route can be generated.
As a result, even if the content of the target route is not appropriate for the user, the target route can be corrected to a more appropriate content by acquiring the vehicle travel locus later.

Further, by appropriately setting the weight information to be given to the target route, the target route can be updated more appropriately.
For example, when many vehicle driving information is acquired by manual driving on the same road, the target route generated by statistically processing the information is said to be optimal for the user. Conceivable.
After that, even if the vehicle traveling information whose traveling position is not appropriate happens to be acquired, the effect of preventing the vehicle from being updated to the inappropriate target route can be expected by reflecting the weight information as described above.

Regarding the weight information given to the target route, an upper limit value is set, and the condition that the content of the new vehicle travel locus acquired after that is not reflected at all in the target route having the weight that reaches the predetermined upper limit value is set. You may add it.
As a result, it is expected that the processing process for newly updating the target route, which is considered to be the most suitable for the user, will be reduced.

In addition, in the area where the approximate curve newly acquired by the operation and the existing target route are adjacent to each other, the operation is corrected for the last section of the calculated approximate curve and smoothly connected to the existing target route. Processing may be performed.
As a result, the newly acquired approximate curve is added to the existing target route, and a longer target route is generated.

Embodiment 3.
Hereinafter, the third embodiment will be described with reference to the drawings.
FIG. 14 is a block diagram showing a configuration of a target route generation system according to the third embodiment.
In FIG. 14, reference numerals 11 to 15 are the same as those in FIG. The target route generation system 200 has a vehicle 300 and a server 400. There are a plurality of vehicles 300, each of which is equipped with a vehicle travel information acquisition unit 11 and a vehicle travel locus generation unit 12, and further, a communication unit 16 for communicating with the server 400.
The server 400 is equipped with a vehicle travel locus dividing unit 13, an approximate curve calculation unit 14, a target route generation unit 15, and a communication unit 17 for communicating with a plurality of vehicles 300.

The processing content until the target route having the configuration of FIG. 14 is generated is the same as the processing shown in the flowchart of FIG.

The difference from the first embodiment in FIG. 14 will be described below.
In the vehicle 300, when the process of generating the vehicle travel locus by the vehicle travel locus generation unit 12 (step S103) is completed, the vehicle travel locus is transferred to the server via the communication unit 16 of the vehicle 300 and the communication unit 17 of the server 400. It is sent to 400.
After that, the server 400 executes the division processing of the vehicle traveling locus, the calculation processing of the approximate curve, and the generation processing of the target route.
Then, the target route generated by the server 400 can be transmitted not only to the vehicle that transmitted the vehicle travel locus but also to all the vehicles 300 equipped with the target route generation system 200.

Further, as a derivative of the configuration shown in FIG. 14, the configuration of the target route generation system shown in FIG. 15 is also possible.
FIG. 15 is a block diagram showing another configuration of the target route generation system according to the third embodiment.
In FIG. 15, reference numerals 11 to 17, 200, 300 and 400 are the same as those in FIG. In FIG. 15, the target route generated by the target route generation unit 15 in the server 400 can be input to the approximate curve calculation unit 14.

According to the configuration of FIG. 15, the approximate curve calculation unit 14 uses both the target route output from the target route generation unit 15 and the vehicle travel locus transmitted from the vehicle 300 via the communication unit 17 to obtain an approximate curve. Can be calculated.
The processing method for calculating the approximate curve using the target route and the vehicle travel locus according to the configuration of FIG. 15 is the same as the processing shown in the flowchart of FIG.
Then, the target route generated by the server 400 can be transmitted not only to the vehicle that transmitted the vehicle travel locus but also to all the vehicles 300 equipped with the target route generation system 200.

According to the third embodiment, the information of the target route generated by the vehicle other than the own vehicle can be shared between the vehicles.
Therefore, even if it is the first road for the own vehicle (a road that has never been manually driven), the target route can be used.

Further, in the configuration of the target route generation system 200 shown in FIG. 15, the travel loci of other vehicles traveling on the same road are used for the update process of the target route.
Since the target route is updated in the server, the information of the target route can be shared between vehicles, and it is appropriate even on a road where the number of acquisitions of vehicle driving information is small for the own vehicle or there is no acquisition history. The own vehicle can acquire the target route that has been corrected to the above contents.

The target route generation device 100 is composed of a processor 101 and a storage device 102, as shown in FIG. 16 as an example of hardware. Although the storage device is not shown, it includes a volatile storage device such as a random access memory and a non-volatile auxiliary storage device such as a flash memory. Further, the auxiliary storage device of the hard disk may be provided instead of the flash memory. The processor 101 executes the program input from the storage device 102. In this case, the program is input from the auxiliary storage device to the processor 101 via the volatile storage device. Further, the processor 101 may output data such as a calculation result to the volatile storage device of the storage device 102, or may store the data in the auxiliary storage device via the volatile storage device.

The present disclosure describes various exemplary embodiments and examples, although the various features, embodiments, and functions described in one or more embodiments are those of a particular embodiment. It is not limited to application, but can be applied to embodiments alone or in various combinations.
Therefore, innumerable variations not exemplified are envisioned within the scope of the techniques disclosed herein. For example, it is assumed that at least one component is modified, added or omitted, and further, at least one component is extracted and combined with the components of other embodiments.

11 Vehicle travel information acquisition unit, 12 Vehicle travel locus generation unit, 13 Vehicle travel locus division unit,
14 Approximate curve calculation unit, 15 target route generation unit, 16 communication unit, 17 communication unit,
21 Actual travel locus, 22 Approximate curve, 23 Target route, 31 Area,
32 target route extraction area, 100 target route generator, 101 processor,
102 storage, 200 target route generation system, 300 vehicles, 400 servers

Claims (8)

Vehicle driving information acquisition unit that acquires vehicle driving information including the above vehicle position information when the vehicle is manually driven,
A vehicle travel locus generation unit that generates a vehicle travel locus in which each position information is arranged in chronological order from the vehicle travel information acquired by this vehicle travel information acquisition unit.
A vehicle travel locus dividing unit that divides a vehicle travel locus generated by this vehicle travel locus generation unit into a plurality of sections based on the above position information.
Based on the vehicle travel locus divided by the vehicle travel locus dividing portion, the approximate curve of the travel locus when the vehicle is manually driven is continuous for each section and at the boundary of the adjacent section. Approximate curve calculation unit that calculates with the end point of the section as the start point of the next section ,
A target route generation device including a target route generation unit that combines approximate curves of each section calculated by the approximate curve calculation unit to generate a target route for vehicle travel. The target route generation system according to claim 1 , wherein the approximate curve calculation unit calculates the approximate curve using a past target route generated by the target route generation unit. Vehicle driving information acquisition unit that acquires vehicle driving information including the above-mentioned vehicle position information when the vehicle is manually driven,
A vehicle travel locus generation unit that generates a vehicle travel locus in which each position information is arranged in chronological order from the vehicle travel information acquired by this vehicle travel information acquisition unit.
A vehicle travel locus dividing unit that divides a vehicle travel locus generated by this vehicle travel locus generation unit into a plurality of sections based on the above position information.
Based on the vehicle travel locus divided by the vehicle travel locus dividing unit, the calculation is made so that the approximate curve of the travel locus when the vehicle is manually traveled is continuous for each section and at the boundary of the adjacent section. Approximate curve calculation unit,
A target route generation unit for generating a target route for vehicle traveling by combining the approximate curves of each section calculated by this approximate curve calculation unit is provided .
The approximate curve calculation unit is a target route generation device characterized in that the approximate curve is calculated by using the past target route generated by the target route generation unit. Weight information is added to the above target route.
The target according to claim 2 or 3 , wherein the approximate curve calculation unit reflects the target path in the calculation of the approximate curve according to the weight information given to the past target path. Route generator. The target according to any one of claims 1 to 4, wherein the vehicle travel locus generation unit selects each position information of the vehicle travel information and then generates the vehicle travel locus. Route generator. The vehicle driving information acquired by the vehicle driving information acquisition unit includes direction information and includes direction information.
The target route generation device according to any one of claims 1 to 5 , wherein the vehicle traveling locus dividing unit divides the section based on the directional information. It is a target route generation system that is installed in vehicles and servers that can communicate with each other and generates target routes for vehicle travel.
The above vehicle
Vehicle driving information acquisition unit that acquires vehicle driving information including the position information of the vehicle when the vehicle is manually driven,
It is provided with a vehicle travel locus generation unit that generates a vehicle travel locus in which each position information is arranged in chronological order from the vehicle travel information acquired by this vehicle travel information acquisition unit.
The vehicle travel locus generated by the vehicle travel locus generator is transmitted to the server, and the vehicle travel locus is transmitted to the server.
The above server is
A vehicle travel locus dividing unit that divides a vehicle travel locus transmitted from the vehicle into a plurality of sections based on the position information.
Based on the vehicle travel locus divided by the vehicle travel locus dividing unit, the calculation is made so that the approximate curve of the travel locus when the vehicle is manually driven is continuous for each section and at the boundary of the adjacent section. Approximate curve calculation unit,
A target route generation unit for generating a target route for vehicle traveling by combining the approximate curves of each section calculated by this approximate curve calculation unit is provided.
The target route generated by the target route generation unit is transmitted to the vehicle, and the target route is transmitted to the vehicle.
The approximate curve calculation unit is a target route generation system characterized by calculating the approximate curve using a past target route generated by the target route generation unit . A plurality of the above vehicles are connected to the above server so as to be able to communicate with each other.
The seventh aspect of claim 7 , wherein the server uses the vehicle travel locus transmitted from one of the vehicles to transmit the target route generated by the target route generation unit to another vehicle. Target route generation system.

Images