JP2010266306A - Method for generating three-dimensional road center line data and apparatus of the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To automatically generate three-dimensional road center line data based on a three-dimensional traveling track. <P>SOLUTION: A control point detecting means detects a control point from a planar traveling track. The planar traveling track as a planar line shape is generated from a traveling track of a position sensor obtained when it travels on a road. A circular arc detecting means detects a circular arc associated with a control point when the planar line shape is generated. A parabola detecting means detects a circular arc approximated to a parabola when a longitudinal line shape is generated. A relaxation curve detecting means detects a relaxation curve connected to the circular arc when the planar line shape is generated. Since a line detecting means detects a line from the detected control point, the circular arc and the relaxation curve, a line for connecting the control point, the circular arc, the relaxation curve and the line can smoothly be generated as a center line of the planar line shape and the longitudinal line shape. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a method and apparatus for creating three-dimensional road centerline data in three-dimensional road data used for a three-dimensional city model using a computer.

A three-dimensional city model has three elements: a polygon with a three-dimensional coordinate value that represents a building, topography, etc., a texture to be pasted on the polygon, and various information that forms a city function, such as road centerline and signal control information. This means data consisting of Since this three-dimensional city model simulates a virtual city or a real city, it is created for use in a simulation field such as a driving simulator or a flood simulator.
In the three-dimensional city model, three-dimensional road data is generated as a part of the configuration. Therefore, road centerline data is generated, and a road is generated along the road centerline data.

  Conventionally, it has been necessary for a human to set a road center based on a road design drawing or a map, and to input a control point of the road center line. This control point determines the basic shape of the three-dimensional road center line, and various parameters of the three-dimensional road center line change by changing the position of the control point.

  Conventionally, it takes time and effort to manually input the control points and parameters. In addition, the road center line that was created due to a mistake in the drawing could be different from the actual one.

  An object of the present invention is to automatically generate three-dimensional road centerline data based on a three-dimensional traveling locus.

  In order to solve the above-mentioned problem, a method for generating three-dimensional road centerline data according to claim 1 is such that a point group of a traveling locus represented by three-dimensional locus data acquired by a position sensor that has moved on a road is a plane alignment and a traveling direction. A control point detection process for detecting, as a control point, a point where two straight line portions located in the middle of a curved portion intersect from a plane traveling locus generated by the locus data generating means for plane alignment with the defining direction of the traveling locus; An arc detection process for detecting an arc of a curved portion formed by the vicinity detected as the control point from the plane traveling locus, and one control point sequentially connecting the control points detected in the control point detection process as an end A relaxation curve detection process for detecting a relaxation curve that smoothly connects two straight lines and a circular arc that is detected by the circular arc detection process in the vicinity of the control point and is not in contact with the straight line; (A) a straight line connecting the control point at the start of the travel locus and the starting point of the arc or relaxation curve in the defining direction is detected, and (b) the end point of the arc or the A straight line connecting the end point of the relaxation curve and the starting point of the arc in the next direction or the starting point of the relaxation curve or the control point is detected. (C) If the detected straight line is not connected to the control point, a straight line detection process in which a) to (b) are repeated.

  In order to solve the above-mentioned problem, the method for generating the 3D road centerline data according to claim 2 is characterized in that the traveling locus point group represented by the 3D locus data acquired by the position sensor moved on the road has a longitudinal linear shape. A control point detection process for detecting, as a control point, a point at which two straight line portions located in the middle of a curved portion intersect from a vertical travel locus generated by the trajectory data generation means for vertical alignment with the defining direction of the travel locus; A parabola detection process in which a parabola of a curved portion formed by the vicinity detected as the control point from the longitudinal traveling locus is detected by approximating with a circular arc, and a point group of the traveling locus is sequentially examined in a defined direction, and (a) traveling Detecting a straight line connecting the control point at the start of the trajectory and the starting point of the arc in the defining direction; and (b) detecting a straight line connecting the end point of the arc and the starting point or control point of the arc in the next direction; ) Until said detected straight line connecting the control points if not connected to the control point, it is characterized in that it has a line detection step of repeating said (a) from (b).

  In order to solve the above-described problem, a 3D road centerline data generating apparatus according to claim 3 is configured such that a point group of a travel locus represented by 3D locus data acquired by a position sensor moved on a road is planar and a travel direction is determined. From the plane linear trajectory data generation means for generating a plane travel trajectory as the definition direction of the travel trajectory, and the plane travel trajectory from the plane linear trajectory data generation means, two straight line portions located in the middle of the curved portion intersect. Control point detecting means for detecting a point as a control point, and arc detecting means for detecting an arc of a curved portion formed by the vicinity detected as the control point from a plane running locus from a locus data generating means for plane alignment The control points detected by the control point detection means are connected to the two straight lines with one control point as an end, and the vicinity of the control point is detected by the arc detection process and is not in contact with the straight line. A relaxation curve detection means for detecting a relaxation curve that smoothly connects the arc and a point group of the travel locus in order in the defined direction, and (a) a control point at the start of the travel locus and a starting point of the arc in the defined direction Alternatively, a straight line connecting the starting point of the relaxation curve is detected, and (b) a straight line connecting the end point of the arc or the end point of the relaxation curve and the starting point of the arc in the next direction or the starting point of the relaxation curve or the control point is detected. c) It has a straight line detecting means for repeating (a) to (b) until the detected straight line is not connected to the control point.

  In order to solve the above-described problem, a three-dimensional road centerline data generating device according to a fourth aspect of the present invention is configured such that the point of the traveling locus represented by the three-dimensional locus data acquired by the position sensor moved on the road is vertically linear and the traveling direction is determined. From the longitudinal linear trajectory data generating means for generating a vertical traveling trajectory as the defining direction of the travel trajectory, and the vertical traveling trajectory from the longitudinal linear trajectory data generating means, two straight line portions located at the middle of the curved portion intersect. A control point detecting means for detecting a point as a control point, a parabola detecting means for approximating a parabola of a curved portion formed by the vicinity detected as the control point from the longitudinal traveling locus by an arc, and traveling in a defined direction The trajectory point group is examined sequentially, (a) a straight line connecting the control point at the start of the travel trajectory and the starting point of the arc in the defined direction is detected, and (b) the end point of the arc and the next direction are detected. A straight line detecting means for detecting a straight line connecting the starting point of the arc or the control point, and (c) a straight line detecting means for repeating (a) to (b) until the detected straight line is not connected to the control point. It is characterized by this.

  According to the three-dimensional road centerline data generation method of claim 1, a control point is detected from a plane travel locus generated as a plane alignment by using a travel locus of a position sensor acquired during road travel, and related to the control point. The arc is detected, the relaxation curve connected to the arc is detected, and the straight line is detected from the detected control point, arc, and relaxation curve. The control line, arc, relaxation curve, and the line connecting the straight lines are the center line. Can be generated smoothly.

  According to the method for generating the three-dimensional road centerline data according to claim 2, a control point is detected from a longitudinal trajectory generated as a vertical alignment from the trajectory of the position sensor acquired during road travel, and Since a parabola approximated by an arc is detected and a straight line is detected from the detected control point and arc, a line connecting these control point, arc and straight line can be smoothly generated as a center line.

  According to the three-dimensional road centerline data generating device of claim 3, the plane linear trajectory data generating means generates a plane driving trajectory by using the traveling trajectory of the position sensor acquired during road driving as a plane alignment, and detects a control point. The means detects a control point from the plane travel locus, the arc detection means detects the arc in relation to the control point, the relaxation curve detection means detects the relaxation curve connected to the arc, and the straight line detection means detects the control point, Detect straight lines from arcs and relaxation curves. A line connecting these control points, arcs, relaxation curves, and straight lines can be smoothly generated as a center line.

  According to the three-dimensional road centerline data generating device of claim 4, the longitudinal linear trajectory is generated by using the linear trajectory data generating means as the vertical trajectory of the position sensor acquired by the trajectory data generating means for vertical alignment. The means detects a control point from the longitudinal traveling locus, the parabola detection means detects a parabola approximated by an arc in relation to the control point, and the straight line detection means detects a straight line from the control point and the arc. A line connecting these control points, arcs, and straight lines can be smoothly generated as a center line.

FIG. 1 is a functional block diagram of an embodiment of the apparatus of the present invention. FIG. 2a is a flowchart illustrating a method for generating a planar alignment. FIG. 2B is a flowchart illustrating a method for generating a vertical alignment. FIG. 3 is a diagram for explaining generation of trajectory data for plane alignment and longitudinal alignment. FIG. 4 is a diagram for explaining generation of planar alignment and longitudinal alignment. FIG. 5 is a diagram for explaining detection of control points. FIG. 6 is a diagram for explaining detection of an arc. FIG. 7 is a diagram for explaining the detection of the relaxation curve. FIG. 8 is a diagram illustrating creation of a relaxation curve. FIG. 9A is a diagram for explaining detection of a straight line in the plane alignment. FIG. 9B is a diagram for explaining detection of a straight line in a longitudinal alignment. FIG. 10 is a diagram for explaining detection of a parabola.

FIG. 1 is a functional block diagram of an embodiment of the apparatus of the present invention, in which 101 is a three-dimensional running locus data storage unit, 102 is a plane linear generation unit, 103 is plane linear locus data generation means, and 104 is a control. Point detection means, 105 is an arc detection means, 106 is a relaxation curve detection means, 107 is a straight line detection means, 108 is a longitudinal linear generation unit, 109 is a longitudinal linear trajectory data generation means, 110 is a control point detection means, and 111 is Parabola detection means, 112 is a straight line detection means, 113 is a three-dimensional road centerline data storage unit, 114 is an image generation unit, and 115 is a monitor.
2a and 2b are flow diagrams for implementing the method of the present invention.

The three-dimensional traveling locus data storage unit 101 stores three-dimensional traveling locus data. This data is obtained by mounting a three-dimensional position sensor such as a GPS on a car (not shown), running on a real road in advance, sampling a travel locus at a predetermined interval, and acquiring the three-dimensional travel locus. Yes, it is assumed that the vehicle traveled on orthogonal coordinates with the vertical direction set to the Z axis.
The plane linear generation unit 102 generates plane linear data from the three-dimensional trajectory data. The longitudinal linear generation unit 108 generates longitudinal linear data from the three-dimensional trajectory data.
First, the operator reads out the three-dimensional traveling locus data from the three-dimensional traveling locus data storage unit 101 and transfers the three-dimensional traveling locus data to the plane linear locus data generating means 103 and the longitudinal linear locus data generating means 109 to obtain the plane alignment and longitudinal alignment data. Trajectory data is generated (FIG. 3). This generation status is displayed on the monitor 115, and the operator can see this screen.

<Generation of planar alignment>
The generation of the plane alignment in the plane alignment generation unit 102 will be described (FIG. 4A).
The plane linear trajectory data generation means 103 converts the trajectory from the three-dimensional trajectory data into a trajectory on the XY plane to generate plane linear trajectory data (P201a in FIG. 2a). At this time, the traveling direction of the position sensor when obtaining the three-dimensional trajectory data is set as the defining direction of the traveling trajectory (FIG. 4).
The control point detection unit 104 executes a control point detection process, detects the control point from the plane linear locus data generated by the plane linear locus data generation unit 103, and obtains the position (P202a in FIG. 2a).
<Detection of control points>
This control point detection process P202a is performed by the control point detection means 104 as follows, for example.
1. An equation of a straight line composed of two points randomly selected from the traveling locus point group in each curved portion selected from the locus data for plane alignment is obtained. The operator specifies the range of the curved portion. (Fig. 5 (a))
2. The minimum distance from the obtained straight line is obtained for all the point groups of the corresponding curved portion. (Fig. 5 (b))
3. The number n of point groups in which the obtained distance is equal to or less than a preset threshold value d is obtained. (Fig. 5 (c))
4). Steps 1 to 3 are repeated N times at an appropriate value, and the straight line when n reaches the maximum value (n _max ) is defined as the detected straight line. (Fig. 5 (d))
5). Steps 1 to 4 are further repeated using a point group obtained by excluding the point group included in the straight line detected in 4 from the point group of the curve portion selected for processing in 1 above. (Fig. 5 (e))
6). By performing steps 1 to 5, a total of two straight lines can be obtained at each curve portion. A point where the two straight lines intersect is defined as a control point, and there is one control point in each curve portion. In addition, two points located at both ends of all the traveling locus point groups are regarded as control points unconditionally.
Note that d and N are determined and adjusted by the operator based on the processing time and detection accuracy.
In the detection of the control points, the processing of 1 to 6 may be performed for each curved portion and repeated until the control points are detected for all the curved portions, or each processing of 1 to 6 may be sequentially performed on all the curved portions. .
The control points for plane alignment are recorded in the three-dimensional road centerline data recording unit 113.

<Arc detection>
Next, the arc detecting means 105 executes an arc detecting process to detect arc parameters (P203a in FIG. 2a).
This arc detection process P203a is performed by the arc detecting means 105 as follows, for example.
1. The operator generates the image data by the image generation unit 114 based on the trajectory data for plane alignment and displays <control by the control point detection unit 104 in the traveling trajectory point group in each curve portion on the screen 115. Point Detection> Three points are randomly selected from the traveling locus point group excluding the point group constituting the two straight lines obtained in “6” in the process P202a. The arc detecting means 105 obtains a circle formula. (Fig. 6 (a))
The selection of these three points may be made automatically by a program.
2. The arc detecting means 105 obtains the minimum distance from the obtained circle for all point groups. (Fig. 6 (b))
3. The arc detecting means 105 obtains the number n of point groups whose obtained distance is equal to or less than a preset threshold value d. (Fig. 6 (c))
4). The arc detecting means 105 repeats steps 1 to 3 N times to obtain a circle when n reaches the maximum value (n _max ). The circle radius at this time is taken as the radius R of the arc and the circle center as M. (Fig. 6 (d))
5). The arc detecting means 105 sets the first point of the n _max point groups used for the circle detected in 4 as the starting point BP of the arc. An angle formed by a line connecting the circle center and the first point and a line connecting the circle center and the last point is defined as a circle center angle θ. (Fig. 6 (d))
6). The arc detection means 105 determines the arc defined by the radius R, the circle center M, the circle center angle θ, and the starting point BP determined in the above 1 to 5 as the finally detected arc, and the control point detection means 104 <Control Point Detection> The control point detected in the process P202a and this arc have a one-to-one relationship.
Note that d and N are determined and adjusted by the operator based on the processing time and detection accuracy.
In the detection of the arc, the processes 1 to 6 may be performed for each curved part and repeated until the arcs are detected for all the curved parts, or the processes 1 to 6 may be sequentially performed for all the curved parts.
The radius R, the circle center M, the circle center angle θ, and the starting point BP for plane alignment are stored in the three-dimensional road centerline data recording unit 113 in association with the control points having the one-to-one relationship with the arc parameters. To be recorded.

<Detection of relaxation curve>
Next, the relaxation curve detection means 106 executes a relaxation curve detection process to detect a relaxation curve parameter (P203a in FIG. 2a).
This relaxation curve detection process P203a is performed by the relaxation curve detection means 106 as follows, for example.
1. In the entire traveling locus point group on the XY plane displayed on the monitor 115, the operator
<Control Point Detection> by Control Point Detection Unit 104 The control points detected in the process P202a are connected to form a straight line group. (Fig. 7 (a))
The straight line group may be created by the relaxation curve detecting means 106 automatically by a program.
2. The operator or the relaxation curve detection means 106 includes a control point IP and two straight lines that end at the control point among the control points and the straight line group in 1 and an arc that has a one-to-one relationship with the control point. Select for C. In addition, let two straight lines be La and Lb in order of the definition direction of a run locus point group. When the selection is made by the relaxation curve detection means 106, for example, it is performed in the order in which the traveling locus point group is defined. (Fig. 7 (b))
3. The relaxation curve detection means 106 obtains the angle τa formed by the intersection of the tangent at the starting point of the arc C and La, and the angle τb formed by the intersection of the tangent at the end of the arc C and Lb. (Fig. 7 (c))
4). The relaxation curve defined by τa and the radius R of the arc C is Ka, and the relaxation curve defined by τb and the radius R of the arc C is Kb. These are the two relaxation curves at each control point. When τa or τb can be regarded as almost zero, it is assumed that there is no corresponding relaxation curve. (FIG. 7D). The relaxation curve detecting means 106 gradually changes the relaxation curves ka and Kb so that the curvature radii of contact with La and Lb are ∞ and the curvature radius connected to the arc C is R (FIG. 8). The relaxation curve is created by a preset program.
In the detection of the relaxation curve, the processing of 1 to 4 is performed for each control point, and it is repeated until it is confirmed that the relaxation curve is not detected or does not exist for all the control points, or each processing of 1 to 4 is sequentially performed for all the control points. You may make it perform.
The angles τa and τb for plane alignment, the radius R of the circular arc C, and the relaxation curves Ka and Kb are recorded as three-dimensional road centerline data in association with the control points having the one-to-one relationship using the relaxation curve parameters. Recorded in the unit 113.

<Detection of straight lines in planar alignment>
Next, the straight line detection means 107 executes a straight line detection process to detect a straight line parameter (P205a in FIG. 2a).
This straight line detection process P205a is performed by the straight line detection means 107 as follows, for example. (Fig. 9a)
1. In the control point detection process P202a, the arc detection process P203a, and the relaxation curve detection process 204a, the operator displays the control points, arcs, and relaxation curves detected along the traveling locus point group on the monitor 115 according to the parameters. Check the running trajectory point cloud sequentially in the defined direction,
(A) A straight line connecting the control point at the start of the travel locus and the starting point of the arc or relaxation curve in the defining direction is detected.
2. (B) A straight line connecting the end point of the arc or the end point of the relaxation curve and the starting point of the arc or the relaxation curve in the next direction is detected.
3. (C) If the detected straight line is not connected to the control point, the above steps (a) to (b) are repeated until the detected straight line is connected to the control point.
The detected straight line for plane alignment is recorded in the three-dimensional road centerline data recording unit 113 in association with these parameters so as to be successively connected to control points, arcs, and relaxation curves.

<Generation of vertical alignment>
Next, generation of vertical alignment in the vertical alignment generation unit 108 will be described (FIG. 4B).
The longitudinal linear trajectory data generation means 109 converts the trajectory from the three-dimensional trajectory data into a trajectory on the ZY plane to generate longitudinal linear trajectory data (P201b in FIG. 2b). At this time, the traveling direction of the position sensor when obtaining the three-dimensional trajectory data is set as the defining direction of the traveling trajectory (FIG. 4).
<Detection of control points>
The control point detection unit 110 executes a control point detection process, detects the control point from the longitudinal linear locus data generated by the longitudinal linear locus data generation unit 109, and obtains the position (P202b in FIG. 2b).
This control point detection process P202b is performed by the control point detection means 110 as follows, for example. The method of the control point detection process P202b is the same as the control point detection process P202a of <Generation of planar alignment>.
1. An equation of a straight line composed of two points randomly selected from the running locus point group in each curve portion selected from the longitudinal linear locus data is obtained. The range of the curved part is specified by the operator. (Fig. 5 (a))
2. The minimum distance from the obtained straight line is obtained for all the point groups of the corresponding curved portion. (Fig. 5 (b))
3. The number n of point groups in which the obtained distance is equal to or less than a preset threshold value d is obtained. (Fig. 5 (c))
4). Steps 1 to 3 are repeated N times at an appropriate value, and the straight line when n reaches the maximum value (n _max ) is defined as the detected straight line. (Fig. 5 (d))
5). Steps 1 to 4 are further repeated using a point group obtained by excluding the point group included in the straight line detected in 4 from the point group of the curve portion selected for processing in 1 above. (Fig. 5 (e))
6). By performing steps 1 to 5, a total of two straight lines can be obtained at each curve portion. A point where the two straight lines intersect is defined as a control point, and there is one control point in each curve portion. In addition, two points located at both ends of all the traveling locus point groups are regarded as control points unconditionally.
Note that d and N are determined and adjusted by the operator based on the processing time and detection accuracy.
In the detection of the control points, the processing of 1 to 6 may be performed for each curved portion and repeated until the control points are detected for all the curved portions, or each processing of 1 to 6 may be sequentially performed on all the curved portions. .
The longitudinal alignment control points are recorded in the three-dimensional road centerline data recording unit 113.

<Detection of parabola>
Next, the parabola detection means 111 executes a parabola detection process, and detects the arc parameter by approximating the parabola of the curved portion formed by the vicinity detected as the control point with an arc (P204b in FIG. 2b). ).
Since the parabola in the vertical alignment can be approximated by an arc, steps 1 to 4 of the <arc detection> process P203a in the plane alignment are executed for the traveling locus point group of each curved portion in the ZY plane. Accordingly, a parabola having the same radius as the detected radius R of the arc is taken as a parabola at each curved portion. (Fig. 10)
This parabola detection process P204b is performed by the parabola detection means 111 as follows, for example.
1. The operator generates the image data by the image generation unit 114 based on the trajectory data for vertical alignment, and on the screen displayed on the monitor 115, the control point detection unit 110 controls the control point detection unit 110 out of the travel trajectory point group in each curve portion. Point Detection> Three points are randomly selected from the traveling locus point group excluding the point group constituting the two straight lines obtained in “6” in the process P202b. The parabola detecting means 111 calculates a circle formula. (Fig. 6 (a))
The selection of these three points may be made automatically by a program.
2. The parabola detecting means 111 obtains the minimum distance from the obtained circle for all point groups. (Fig. 6 (b))
3. The parabola detecting unit 111 obtains the number n of point groups in which the obtained distance is equal to or less than a preset threshold value d. (Fig. 6 (c))
4). The parabola detecting means 111 repeats steps 1 to 3 N times to obtain a circle when n reaches the maximum value (n _max ), and the circle radius at this time is taken as the radius R of the arc and M as the circle center. (Fig. 6 (d))
5). The parabola detecting means 111 sets the first point of the n _max point groups used for the circle detected in 4 as the starting point BP of the arc. An angle formed by a line connecting the circle center and the first point and a line connecting the circle center and the last point is defined as a circle center angle θ. (Fig. 6 (d))
6). The parabola detection unit 111 sets the arc defined by the radius R, the circle center M, the circle center angle θ, and the starting point BP determined in the above 1 to 5 as a finally detected arc, and is performed by the control point detection unit 110. <Detection of Control Point> The control point detected in the process P202b and this arc have a one-to-one relationship.
Note that d and N are determined and adjusted by the operator based on the processing time and detection accuracy.
In the detection of the arc, the processes 1 to 6 may be performed for each curved part and repeated until the arcs are detected for all the curved parts, or the processes 1 to 6 may be sequentially performed for all the curved parts.
The longitudinal linear radius R, the circle center M, the circle center angle θ, and the starting point BP are stored in the three-dimensional road centerline data recording unit 113 in association with the control points having the one-to-one relationship with the arc parameters. To be recorded.

<Detection of straight lines in longitudinal alignment>
Next, the straight line detection means 112 executes a straight line detection process to detect a straight line parameter (P205b in FIG. 2b).
This straight line detection process P205b is performed by the straight line detection means 112 as follows, for example. (Fig. 9b)
1. In the control point detection process P202b and the parabola detection process P204b, the operator defines the point group of the traveling locus in which the control points and the arc detected along the traveling locus point group are displayed on the monitor 115 according to the parameters. Check the direction sequentially,
(A) A straight line connecting the control point at the start end of the travel locus and the starting point of the arc in the definition direction is detected.
2. (B) A straight line connecting the end point of the arc and the starting point or control point of the arc in the next direction is detected.
3. (C) If the detected straight line is not connected to the control point, the above steps (a) to (b) are repeated until the detected straight line is connected to the control point.
The detected vertical straight line is recorded in the three-dimensional road centerline data recording unit 113 in association with these parameters so as to be successively connected to the control points and the arc.

101 ... three-dimensional traveling locus data storage unit,
102 ... a plane linear generator,
103 ... plane linear locus data generating means,
107: Straight line detecting means,
104: Control point detection means,
105... Arc detection means,
106: relaxation curve detection means,
108: Longitudinal linear generator,
109 ... Longitudinal linear trajectory data generation means,
112 ... Straight line detection means,
110 ... Control point detection means,
111 ... Parabolic detection means,
113 ... 3D road centerline data storage unit,
114... The image generation unit,
115: Monitor.

Claims (4)

A plane running trajectory generated by the trajectory data generation means for plane alignment with the point group of the running trajectory represented by the three-dimensional trajectory data acquired by the position sensor moved on the road as the plane alignment and the running direction as the definition direction of the running trajectory From the control point detection process of detecting, as a control point, a point where two straight line portions located in the middle of the curved portion intersect,
An arc detection process for detecting an arc of a curved portion formed by the vicinity detected as the control point from the plane traveling locus;
Smoothly connect two straight lines having one control point connected sequentially to the control points detected in the control point detection process and an arc not detected in contact with the straight line in the vicinity of the control point. A relaxation curve detection process for detecting a relaxation curve;
Sequentially examine the point cloud of the running locus in the defined direction,
(A) detecting a straight line connecting the control point at the start of the travel locus and the starting point of the arc or relaxation curve in the defining direction;
(B) detecting a straight line connecting the end point of the arc or the end point of the relaxation curve and the starting point of the arc in the next direction or the starting point or control point of the relaxation curve;
(C) If the detected straight line is not connected to a control point, a straight line detection process is repeated to repeat (a) to (b) until the control point is connected. Method. Longitudinal travel trajectory generated by trajectory data generation means for longitudinal alignment with the trajectory point group represented by the three-dimensional trajectory data acquired by the position sensor moved on the road as the longitudinal alignment and the travel direction as the defining direction of the travel trajectory From the control point detection process of detecting, as a control point, a point where two straight line portions located in the middle of the curved portion intersect,
A parabola detection process for detecting a parabola of a curved portion formed by the vicinity detected as the control point from the longitudinal traveling locus by approximating with a circular arc,
Sequentially examine the point cloud of the running locus in the defined direction,
(A) detecting a straight line connecting the control point at the start of the travel locus and the starting point of the arc in the defined direction;
(B) detecting a straight line connecting the end point of the arc and the starting point or control point of the arc in the next direction;
(C) If the detected straight line is not connected to a control point, a straight line detection process is repeated to repeat (a) to (b) until the control point is connected. Method. Trajectory data generation means for plane alignment that generates a plane travel locus with the plane direction of the travel locus point cloud represented by the three-dimensional locus data acquired by the position sensor moved on the road as the plane alignment,
Control point detecting means for detecting, as a control point, a point where two straight line portions located in the middle of the curved line portion intersect from the plane running locus from the locus data generating means for plane alignment;
Arc detecting means for detecting an arc of a curved portion formed by the vicinity detected as the control point from a plane running locus from the locus data generating means for plane alignment;
Smoothly connect two straight lines having one control point as an end sequentially connecting the control points detected by the control point detecting means and an arc not detected in contact with the straight line in the vicinity of the control point. A relaxation curve detecting means for detecting a relaxation curve;
Sequentially examine the point cloud of the running locus in the defined direction,
(A) detecting a straight line connecting the control point at the start of the travel locus and the starting point of the arc or relaxation curve in the defining direction;
(B) detecting a straight line connecting the end point of the arc or the end point of the relaxation curve and the starting point of the arc in the next direction or the starting point or control point of the relaxation curve;
(C) If the detected straight line is not connected to a control point, straight line detection means that repeats (a) to (b) until the control point is connected is generated. apparatus. Trajectory linear trajectory data generation means for generating a vertical travel trajectory with the travel trajectory point cloud represented by the three-dimensional trajectory data acquired by the position sensor moved on the road as the vertical trajectory and the travel direction as the travel trajectory definition direction;
A control point detecting means for detecting, as a control point, a point where two straight line portions located in the middle of the curved portion intersect from the longitudinal traveling locus from the trajectory data generating means for vertical alignment;
A parabola detecting means for detecting a parabola of a curved portion formed by the vicinity detected as the control point from the longitudinal traveling locus by approximating with a circular arc;
Sequentially examine the point cloud of the running locus in the defined direction,
(A) detecting a straight line connecting the control point at the start of the travel locus and the starting point of the arc in the defined direction;
(B) detecting a straight line connecting the end point of the arc and the starting point or control point of the arc in the next direction;
(C) If the detected straight line is not connected to a control point, straight line detection means that repeats (a) to (b) until the control point is connected is generated. apparatus.

Images