Travel track generation device, travel track generation method, and travel track generation program

Abstract

To reduce an error which is generated in travel track data of a mobile when dividing the travel track data into a plurality of regions and correcting them.SOLUTION: A travel track generation device sets a plurality of regions positioned in locations along travel track data and including overlapped regions with adjacent regions with respect to the travel track data of each of a plurality of mobiles. The travel track generation device corrects the travel track data which are captured in each of the plurality of regions which are set on the basis of a peripheral target object captured in the travel track data of each of the plurality of mobiles. The travel track generation device corrects the travel track data of each of the plurality of mobiles by correcting a position or an inclination of at least one region of a pair, regarding each pair of adjacent regions set to the travel track data of each of the plurality of mobiles, in such a manner that a mobile in travel track data captured in an overlapped region in a first region of a pair coincides with a mobile in travel track data captured in an overlapped region in a second region of the pair.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a traveling locus generation device, a traveling locus generation method, and a traveling locus generation program. [Background technology]

[0002] Patent Document 1 discloses a trajectory analysis device that accurately identifies the relative position of multiple movement trajectories. The trajectory analysis device disclosed in Patent Document 1 generates trajectories for each of multiple terminal devices based on movement information that indicates, over time, the displacement from the position at the start of movement and the deflection angle from the orientation at the start of movement for each of the multiple terminal devices. The trajectory analysis device then collects intensity information that indicates, over time, the strength of signals arriving at each terminal device from a signal source. Based on the temporal change in the signal strength indicated by the intensity information collected for each terminal device, the trajectory analysis device estimates a reference time indicating the time when each terminal device passed a predetermined reference point. From the trajectories generated for each terminal device, the trajectory analysis device extracts, as a reference range, the range of movement of the terminal device during a predetermined period including the reference time corresponding to the terminal device. The trajectory analysis device then identifies the relative positions of the multiple trajectories by comparing the topological features of the reference range extracted from the trajectories of each terminal device.

[0003] Furthermore, Patent Document 2 discloses an autonomous mobile system that detects overlapping travel areas, where autonomous mobile devices have traveled in an overlapping manner, by detecting areas where trajectories constituting a corrected travel trajectory are close to each other. This autonomous mobile system selects a plurality of relative coordinates on the travel trajectories in the overlapping travel area, calculates a first relative positional relationship between each of the selected relative coordinates by comparing environmental information associated with each of the selected relative coordinates, calculates a second error evaluation value based on both the calculated first relative positional relationship and the absolute coordinates, and corrects the travel trajectory by transforming the travel trajectory of the autonomous mobile device obtained from the relative coordinates or the travel trajectory that has undergone the first correction based on the second error evaluation value. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-217802 [Patent Document 2] Patent No. 5930346 Summary of the Invention [Problem to be solved by the invention]

[0005] When generating travel trajectory data by accumulating the amount of change in position and the amount of change in attitude of a moving body, as the travel trajectory becomes longer, errors occur in the estimation results of the position and attitude of the moving body, and these errors are reflected in the travel trajectory data. Therefore, when generating long travel trajectory data by accumulating the amount of change in position and attitude of a moving body, distortion occurs in the shape of the travel trajectory data.

[0006] Therefore, for example, when generating travel locus data for a long travel locus, if a technique such as that disclosed in Patent Document 1 is used, distortion occurs in the travel locus data.

[0007] On the other hand, in order to improve the estimation accuracy of the position and attitude of the moving object in each area included in the traveling trajectory data, it is conceivable to divide the traveling trajectory data into small areas, generate traveling trajectory data for each small area, and later integrate the data. However, in this case, inconsistencies (e.g., positional deviations) in the position and attitude of the moving object may occur between the divided small areas. For this reason, for example, when using a technology such as that disclosed in Patent Document 2, inconsistencies in the position and attitude of the moving object may occur between areas, and errors may occur in the traveling trajectory data.

[0008] The present disclosure provides a driving trajectory generation device, a driving trajectory generation method, and a driving trajectory generation program that can reduce errors that occur in driving trajectory data when dividing the driving trajectory data of a moving body into multiple regions and correcting it. [Means for solving the problem]

[0009] A first aspect of the present disclosure is a driving trajectory generation device including: a setting unit that sets, for driving trajectory data of each of a plurality of moving bodies, a plurality of regions that are located along the driving trajectory data and that have overlapping regions with adjacent regions; a correction unit that corrects the driving trajectory data that appears in each of the set plurality of regions based on surrounding objects that appear in the driving trajectory data of each of the plurality of moving bodies; and a generation unit that generates driving trajectory data of each of the plurality of moving bodies by correcting the position or tilt of at least one region of each pair of adjacent regions that are set in the driving trajectory data of each of the plurality of moving bodies so that a moving body in the driving trajectory data that appears in the overlapping region in a first region of the pair matches a moving body in the driving trajectory data that appears in the overlapping region in a second region of the pair.

[0010] In a second aspect, in the driving trajectory generation device of the first aspect, the generation unit corrects the position or tilt of at least one of the areas in the pair so that, for each pair of areas, the moving body in the driving trajectory data that appears in the overlapping area in the first area of ​​the pair matches the moving body in the driving trajectory data that appears in the overlapping area in the second area of ​​the pair, then calculates a residual between the position of the moving body that appears in the overlapping area in the first area and the position of the moving body that appears in the overlapping area in the second area, calculates a value that increases as the residual decreases as a moving body position reliability that represents the reliability of the moving body's position information, and excludes, from the driving trajectory data of each of the multiple moving bodies, driving trajectory data for which the moving body position reliability is below a predetermined threshold.

[0011] In a third aspect, in the driving trajectory generation device of the second aspect, the generation unit excludes, from the driving trajectory data of each of the multiple moving bodies, driving trajectory data for which the moving body position reliability is equal to or lower than a predetermined threshold, and then corrects the position or inclination of at least one of the areas in each pair of areas to correct the driving trajectory data of each of the multiple moving bodies, thereby generating driving trajectory data of each of the multiple moving bodies.

[0012] In a fourth aspect, in the driving trajectory generation device of the first aspect, the generation unit corrects the position or tilt of at least one of the areas in the pair so that, for each pair of areas, the moving body in the driving trajectory data that appears in the overlapping area in the first area of ​​the pair matches the moving body in the driving trajectory data that appears in the overlapping area in the second area of ​​the pair, then calculates a residual between the position of the moving body that appears in the overlapping area in the first area and the position of the moving body that appears in the overlapping area in the second area, calculates a value that increases as the residual decreases as an area reliability that represents the reliability of the area pair, and excludes, from the driving trajectory data of each of the multiple moving bodies, area pairs for which the area reliability is below a predetermined threshold.

[0013] In a fifth aspect, in the driving trajectory generation device of the fourth aspect, the generation unit excludes, from the driving trajectory data of each of the multiple moving bodies, pairs of areas whose area reliability is below a predetermined threshold, and then corrects the position or inclination of at least one of the areas in each pair of areas to correct the driving trajectory data of each of the multiple moving bodies, thereby generating driving trajectory data of each of the multiple moving bodies.

[0014] A sixth aspect is a driving trajectory generation method in which a computer executes a process to generate driving trajectory data for each of a plurality of moving bodies, by setting, for driving trajectory data for each of a plurality of moving bodies, a plurality of regions that are located along the driving trajectory data and that have overlapping regions with adjacent regions, correcting the driving trajectory data that appears in each of the set plurality of regions based on surrounding objects that appear in the driving trajectory data for each of the plurality of moving bodies, and correcting the position or tilt of at least one region of the pair so that, for each pair of adjacent regions set in the driving trajectory data for each of the plurality of moving bodies, a moving body in the driving trajectory data that appears in the overlapping region in a first region of the pair matches a moving body in the driving trajectory data that appears in the overlapping region in a second region of the pair, thereby correcting the driving trajectory data for each of the plurality of moving bodies.

[0015] A seventh aspect is a driving trajectory generation program that causes a computer to function as: a setting unit that sets, for driving trajectory data of each of a plurality of moving bodies, a plurality of regions that are located along the driving trajectory data and that have overlapping regions with adjacent regions; a correction unit that corrects the driving trajectory data that appears in each of the set plurality of regions based on surrounding objects that appear in the driving trajectory data of each of the plurality of moving bodies; and a generation unit that generates driving trajectory data of each of a plurality of moving bodies by correcting the position or tilt of at least one region of each pair of adjacent regions set in the driving trajectory data of each of the plurality of moving bodies so that the moving body in the driving trajectory data that appears in the overlapping region in a first region of the pair matches the moving body in the driving trajectory data that appears in the overlapping region in a second region of the pair. [Effects of the Invention]

[0016] According to the present disclosure, when the travel trajectory data of a moving object is divided into a plurality of regions and corrected, an effect is obtained in that errors occurring in the travel trajectory data can be reduced. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 1 is a diagram illustrating an example of a traveling locus generation system according to a first embodiment. [Figure 2] 1 is a block diagram showing an example of a functional configuration of a running locus generation device according to a first embodiment. [Figure 3] 1 is a block diagram showing an example of the configuration of a computer that functions as a running locus generation device according to a first embodiment. FIG. [Figure 4] FIG. 2 is a diagram for explaining a vehicle travel path and surrounding environment information. [Figure 5] FIG. 2 is a diagram for explaining travel locus data. [Figure 6] FIG. 2 is a diagram for explaining travel trajectory data and targets installed on a travel path. [Figure 7]10A and 10B are diagrams for explaining positional deviation and shape error; [Figure 8] 10 is a diagram showing an example in which a plurality of travel trajectory data are divided into small regions and the relative positions of the travel trajectory data are corrected using a common target that exists in each small region. FIG. [Figure 9] FIG. 10 is a diagram for explaining relative position correction of a plurality of travel locus data; [Figure 10] FIG. 10 is a diagram showing an example of a result of relative position correction of travel locus data within a small region. [Figure 11] FIG. 10 is a diagram for explaining residuals when vehicle positions do not match perfectly. [Figure 12] 10A and 10B are diagrams illustrating an example of a result of relative position correction of travel locus data in each small area. [Figure 13] 10A and 10B are diagrams illustrating an example of a result of relative position correction of travel locus data in each small area. [Figure 14] 10A and 10B are diagrams illustrating an example of a result of relative position correction of travel locus data in each small area. [Figure 15] 5 is a flowchart showing an example of a processing flow of a traveling locus generation program according to the first embodiment. [Figure 16] 10 is a flowchart showing an example of the processing flow of a traveling locus generation program according to the second embodiment. [Figure 17] FIG. 10 is a diagram showing the results of an example. DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, embodiments of the technology of the present disclosure will be described in detail with reference to the drawings. Note that in the present embodiment, components and processes that perform the same actions and functions are assigned the same reference numerals throughout the drawings, and duplicated descriptions may be omitted as appropriate. Each drawing is merely a schematic illustration to allow a sufficient understanding of the technology of the present disclosure. Therefore, the technology of the present disclosure is not limited to only the illustrated examples. Furthermore, in the present embodiment, descriptions of configurations that are not directly related to the present invention or well-known configurations may be omitted.

[0019] [First embodiment] Fig. 1 is a diagram showing an example of a running locus generation system 10 according to a first embodiment. As shown in Fig. 1, the running locus generation system 10 includes a vehicle 12, which is an example of a moving body, and a running locus generation device 14. The vehicle 12 and the running locus generation device 14 are connected to each other so as to be able to communicate with each other via communication means 15 such as the Internet.

[0020] The vehicle 12 is equipped with various sensors such as an on-board camera 120, a gyro sensor 122, and a GPS device (not shown).

[0021] The vehicle-mounted camera 120 captures images of the surroundings of the vehicle 12. In this way, an image of the surroundings of the vehicle 12 is acquired.

[0022] The gyro sensor 122 successively detects the angular velocity of the vehicle 12 .

[0023] A GPS device (not shown) successively detects the position of the vehicle 12 by exchanging data with the positioning satellites 124 .

[0024] The image obtained by the on-board camera 120, the angular velocity obtained by the gyro sensor 122, and the position obtained by a GPS device (not shown) are transmitted to the running trajectory generation device 14 via the communication means 15.

[0025] Although only one vehicle 12 is shown in FIG. 1, in reality, multiple vehicles exist, and the driving trajectory generating device 14 acquires various data transmitted from each of the multiple different vehicles and stores it in the data storage unit 140 described later.

[0026] Fig. 2 is a block diagram showing an example of the functional configuration of the running locus generation device 14 according to the first embodiment. As shown in Fig. 2, the running locus generation device 14 functionally includes a data storage unit 140, a locus creation unit 142, a running locus data storage unit 144, a setting unit 146, a correction unit 148, a generation unit 150, and a corrected running locus data storage unit 152.

[0027] 3, the running locus generation device 14 is realized by a computer 50 including a CPU 51, a memory 52 as a temporary storage area, and a non-volatile storage unit 53. The computer 50 of the running locus generation device 14 also includes an input / output interface (I / F) 54 to which external devices, output devices, etc. are connected, and a read / write (R / W) unit 55 that controls reading and writing of data from and to a recording medium 59. The computer also includes a network I / F 56 that is connected to a network such as the Internet. The CPU 51, memory 52, storage unit 53, input / output I / F 54, R / W unit 55, and network I / F 56 are connected to one another via a bus 57.

[0028] The storage unit 53 can be realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc. The storage unit 53 as a storage medium stores a program for causing the computer to function. The CPU 51 reads the program from the storage unit 53, loads it into the memory 52, and sequentially executes the processes contained in the program.

[0029] The data storage unit 140 stores images at each time obtained by the onboard camera 120 mounted on each of the multiple vehicles 12, the angular velocity at each time obtained by the gyro sensor 122, and the position of the vehicle 12 at each time obtained by a GPS device (not shown).

[0030] The trajectory creation unit 142 uses a known method to generate travel trajectory data for each of the plurality of vehicles 12 based on the various data stored in the data storage unit 140. Then, the trajectory creation unit 142 stores the plurality of travel trajectory data in the travel trajectory data storage unit 144.

[0031] Fig. 4 is a diagram for explaining the travel trajectory of the vehicle 12 and surrounding environment information. Fig. 4 is a diagram showing the road traveled by the vehicle 12 as seen from above. Fig. 4 illustrates the travel trajectory r of the vehicle 12, roadside landmarks B1 (for example, signs, etc.), and road surface landmarks B2 (for example, marking lines, etc.).

[0032] The surrounding environment information including information on roadside landmarks B1 (e.g., signs, etc.) and information on road surface landmarks B2 (e.g., lane markings, etc.) is acquired by various sensors including, for example, the in-vehicle camera 120. In addition, the surrounding environment information is linked to the vehicle's travel trajectory r.

[0033] Fig. 5 is a diagram for explaining the traveling trajectory data created by the trajectory creation unit 142. Fig. 5(A) is a diagram of a road having multiple lanes viewed from above, Fig. 5(B) is a diagram showing the vehicle position X at each time in the traveling trajectory data, Fig. 5(C) is a diagram in which multiple traveling trajectory data are superimposed on a road having multiple lanes, Fig. 5(D) is a diagram showing only the traveling trajectory data of Fig. 5(C), and Fig. 5(E) is a diagram showing an example of traveling trajectory data including positional deviations.

[0034] As shown in Figure 5(B), the position of the vehicle 12 at each time can be identified based on, for example, surrounding environment information. It can be seen that the traveling trajectory data shown in Figure 5(E) has no error, and the shape of the traveling trajectory data is distorted. Therefore, the traveling trajectory data shown in Figure 5(E) needs to be corrected.

[0035] FIG. 6 is a diagram for explaining driving trajectory data and targets installed on a road. In FIG. 6(A), multiple driving trajectory data (represented by arrows in FIG. 6) are superimposed on the road, and roadside targets B1-1 and B1-2 are depicted. In FIG. 6(B), the positional relationship between the driving trajectory data and the surrounding roadside targets B1-1 and B1-2 is correctly reproduced. On the other hand, in FIG. 6(C), the positional relationship between the driving trajectory data and the surrounding roadside targets B1-1 and B1-2 and road surface target B2 is incorrect, and the driving trajectory data needs to be corrected.

[0036] Specifically, when correcting the driving trajectory data as shown in FIG. 6(C), the relative position and orientation of a plurality of driving trajectory data acquired on the same road are corrected using surrounding environment information that includes information on common targets (e.g., roadside targets represented by roadside targets B1-1 and B1-2 or lane markings represented by road surface target B2). As a result, for example, driving trajectory data as shown in FIG. 6(D) is obtained. Note that the correction of the relative position and orientation of a plurality of driving trajectory data is also simply referred to as "relative position correction" below.

[0037] Two types of errors can occur in travel trajectory data: positional deviation and shape error. Fig. 7 is a diagram for explaining positional deviation and shape error. Fig. 7(A) is a diagram showing an example of correct travel trajectory data in which neither positional deviation nor shape error has occurred. Fig. 7(B) is a diagram showing an example of travel trajectory data in which a shape error has occurred. Fig. 7(C) is a diagram showing an example of travel trajectory data in which both positional deviation and shape error have occurred. Fig. 7(D) is a diagram showing an example of travel trajectory data in which at least one of positional deviation and shape error has been subjected to relative position correction.

[0038] As shown in Fig. 7, when performing relative position correction on multiple pieces of driving locus data, if the section of the driving locus data is long, sufficient accuracy of relative position correction may not be obtained due to shape errors in the driving locus data. For this reason, it is conceivable to improve the accuracy of relative position correction on multiple pieces of driving locus data in each small area by dividing the multiple pieces of driving locus data into small areas along the driving locus and performing relative position correction on the multiple pieces of driving locus data in each small area.

[0039] FIG. 8 is a diagram showing an example in which a plurality of travel trajectory data are divided into small regions, and the relative position of the travel trajectory data is corrected for each small region using a common target that exists within the small region.

[0040] FIG. 8(A) shows an example in which multiple pieces of travel trajectory data are divided into small regions. FIG. 8(B) shows multiple pieces of travel trajectory data before relative position correction in bold. When relative position correction is performed on the multiple pieces of travel trajectory data for each small region for the multiple pieces of travel trajectory data in FIG. 8(B), correction results such as those shown in FIG. 8(C) are obtained. As shown in FIG. 8(C), by performing relative position correction on the multiple pieces of travel trajectory data within the divided small regions, the accuracy of the relative positions and relative orientations of the multiple pieces of travel trajectory data within each small region is improved. On the other hand, as shown in FIG. 8(C), the relative positions and relative orientations of the multiple pieces of travel trajectory data between small regions (for example, between a certain small region and an adjacent small region) are not consistent, resulting in a deterioration in accuracy.

[0041] Therefore, in this embodiment, when dividing a plurality of pieces of traveling trajectory data into small regions, a plurality of small regions that overlap with adjacent small regions are set, and the position or tilt of the small region is corrected so that the positions of the vehicles in the traveling trajectory data captured in the overlapping regions match, thereby performing relative position correction on the plurality of pieces of traveling trajectory data. This makes it possible to reduce the occurrence of a situation in which the plurality of pieces of traveling trajectory data are inconsistent between small regions (for example, between a certain small region and an adjacent small region). As a result, it becomes possible to correct the plurality of pieces of traveling trajectory data with high accuracy.

[0042] The setting unit 146 sets a plurality of small regions that are located along the travel trajectory data of each of the plurality of vehicles 12 and that overlap with adjacent small regions, for the travel trajectory data of each of the plurality of vehicles 12. Note that a small region is an example of a region in the present disclosure.

[0043] FIG. 9 is a diagram for explaining relative position correction of a plurality of travel locus data in this embodiment.

[0044] 9A, the setting unit 146 sets a plurality of small regions l, m, n, and o for a plurality of pieces of travel locus data r. When setting the plurality of small regions l, m, n, and o, the setting unit 146 sets adjacent small regions and overlapping regions Dl,m ,D m,n ,D n,o A plurality of small regions l, m, n, o are set so that

[0045] The correction unit 148 corrects the traveling trajectory data captured in each of the plurality of set small areas, based on the peripheral targets captured in the traveling trajectory data of each of the plurality of vehicles 12. Specifically, as shown in Fig. 6 above, the correction unit 148 corrects the traveling trajectory data for each small area, based on the peripheral targets captured in the small area.

[0046] For each pair of adjacent small areas set in the driving trajectory data of each of a plurality of vehicles, the generation unit 150 corrects the position or inclination of at least one of the small areas in the pair so that the vehicle in the driving trajectory data that is captured in the overlapping area in the first small area of ​​the pair matches the vehicle in the driving trajectory data that is captured in the overlapping area in the second area of ​​the pair.

[0047] In Fig. 9(A), a small region m contains a set of vehicle positions {p m n,i,t} is shown, and the set of vehicle positions {p n m,i,t} is shown. Note that m and n are indexes for identifying small regions, i is an index for identifying the travel locus data r, and t is a time. m n,i,t is the overlapping region D within the small region m m,n represents the position of the vehicle at time t on the travel path data i shown in n m is the overlapping region D within the small region n n,m represents the position of the vehicle at time t on the driving trajectory data i captured in the image.

[0048] For example, as shown in FIG. 9B, for a pair of adjacent small regions m and n set in a plurality of travel trajectory data, the generation unit 150 generates an overlapping region D m,n The position p of the vehicle in the driving trajectory data i is m n,i,tand an overlapping region D in the second region n of the pair m, n. n,m The position p of the vehicle in the driving trajectory data i is n m,i,t The position or inclination of at least one of the pair m and n of small areas is corrected so that the position and inclination of the pair m and n of small areas match. As a result, for example, as shown in Fig. 9(C), by correcting the position and inclination of the pair m and n of small areas, multiple pieces of running trajectory data are also appropriately corrected between the small areas.

[0049] The generation unit 150 performs the above-described process for each of the plurality of pairs and corrects the travel trajectory data of each of the plurality of vehicles 12. This makes it possible to appropriately generate the travel trajectory data of each of the plurality of vehicles 12. For example, by performing the above-described process, the relative position correction result of the travel trajectory data in the small area shown in FIG. 10(A) becomes the relative position correction result shown in FIG. 10(B).

[0050] Here, the relative position correction of the travel locus data will be described in more detail.

[0051] A set of small areas set in the travel trajectory data is defined as M, and a set of small areas adjacent to a small area m∈M (hereinafter simply referred to as adjacent small areas) is defined as N. m = {n|Subregion m and subregion n have overlapping regions, n∈M}. Also, subregion m and subregion n∈N m The set of driving trajectory data passing through the overlapping area between m,n Let's say.

[0052] Driving trajectory data i ∈R m,n The set of travel trajectory points (or vehicle positions) within the overlapping area of ​​the small area m for m n,i,t}, and the set of travel trajectory points (or vehicle positions) in the overlapping area of ​​small area n is denoted by {p n m,i,t}, and the corresponding point pair (p m n,i,t ,p n m,i,t ) set P mn Here, t is the time when the driving trajectory points are acquired. m n,i,t ,p n m,i,t ) point p m n,i,t and p n m,i,t The same driving i Although these are driving trajectory points at the same time t, the coordinates of small area m and small area n generally do not match because the relative position corrections of multiple driving trajectory data are performed individually for small area m and small area n. Here, the set of corresponding point pairs for all adjacent small areas of small area m is defined as follows:

number

[0053] where P m In order to correct the average position between the target sub-region and its adjacent sub-regions, the driving trajectory points {p i |(p i ,q i )∈P m}, the corresponding point pair (p i ,p i ) and P m Add to the driving trajectory point p i =(x i ,y i ) in a homogeneous coordinate system is expressed as [x i ,y i ,1] T In the homogeneous coordinate system, the group of points {p i |(p i ,q i )∈P m} and the point group {q i |(p i ,q i )∈P m}, and the coordinate transformation matrix T m where S is the rotation matrix and t is the translation vector.

[0054]

number

[0055] The set of driving trajectories passing through the small region m is R m and the travel trajectory point p t m,i =(x t m,i ,y t m,i ) in a homogeneous coordinate system is expressed as [x t m,i ,y t m,i ,1] T Then, T m In this coordinate transformation, the travel trajectory R m The relative positional relationship within each small area does not change, and the results of the relative position correction performed within each small area are maintained.

[0056] As mentioned above, even if the position or inclination of each pair of small areas is corrected so that the position of the vehicle reflected in the overlapping area in the first small area (the above-mentioned driving trajectory point) matches the position of the vehicle reflected in the overlapping area in the second small area (the above-mentioned driving trajectory point), the vehicle positions generally do not match perfectly.

[0057] FIG. 11 is a diagram for explaining residuals when vehicle positions do not completely match. As shown in FIG. 11, consider a case where the position and tilt of at least one of small regions m and n is corrected so that the vehicle positions (represented by black circles and white circles in FIG. 11) in a pair of small regions m and n match. In this case, the vehicle position in small region m (black circle in FIG. 11) in the travel trajectory data for i=3 in FIG. 11 must match the vehicle position in small region n (white circle in FIG. 11). However, in the example of FIG. 11, the vehicle position in small region m (black circle) in the travel trajectory data for i=3 does not match the vehicle position in small region n (white circle), resulting in the occurrence of residuals. On the other hand, in the example of FIG. 11, almost no residuals occur in the travel trajectory data for i=1, 2, 4, 5, and 6.

[0058] Therefore, by calculating the residual error of the vehicle position after the relative position correction between the small areas, it is possible to determine the reliability of the relative position correction of the traveling locus data within each small area.

[0059] Therefore, for each pair of small areas, the generation unit 150 corrects the position or tilt of the small areas so that the position of the vehicle in the traveling trajectory data captured in the overlapping area in the first area of ​​the pair coincides with the position of the vehicle in the traveling trajectory data captured in the overlapping area in the second area of ​​the pair, and then calculates the residual between the position of the vehicle captured in the overlapping area in the first area and the position of the vehicle captured in the overlapping area in the second area. Note that the residual is calculated for each piece of traveling trajectory data.

[0060] Next, the generation unit 150 calculates a value according to the residual (for example, the average or median of the residual) as the vehicle position reliability, which indicates the reliability of the vehicle position information. Note that the larger the residual, the smaller the vehicle position reliability, and vice versa.

[0061] Next, the generation unit 150 excludes (or invalidates) the traveling trajectory data for which the vehicle position reliability is equal to or less than a predetermined threshold value from the traveling trajectory data of each of the plurality of vehicles 12. For example, the generation unit 150 excludes the traveling trajectory data for i=3 from the traveling trajectory data shown in FIG.

[0062] The generation unit 150 excludes (or invalidates) the travel trajectory data whose vehicle position reliability is equal to or lower than a predetermined threshold from among the plurality of travel trajectory data, and then performs the above-described process again to correct the position or tilt of the small area and generate a plurality of travel trajectory data.The generation unit 150 then stores the corrected plurality of travel trajectory data in the corrected travel trajectory data storage unit 152.This excludes data with low reliability, and enables the plurality of travel trajectory data to be corrected with higher accuracy.

[0063] 12 to 14 are diagrams showing examples of the relative position correction results of the travel locus data in each small area.

[0064] 12 to 14 show examples of relative position correction results for four driving trajectory data of the same lane. FIG. 12 shows an example in which relative position correction within each small area m, n has been performed correctly. In FIG. 12, when small area m and small area n are superimposed within the overlapping area, the black target and white target are in appropriate positions. In this way, when relative position correction within each small area m, n has been performed correctly, relative position correction between small areas will also be performed correctly.

[0065] On the other hand, FIG. 13 shows an example in which relative position correction within each small region is not performed correctly. In the example of FIG. 13, since relative position correction within each small region is not performed correctly, residual errors occur when correcting relative positions between small regions. In this example, the relative position correction results within each small region appear correct at first glance, but a deviation of one period occurs in some of the travel trajectory data. In such a case, it can be determined that the reliability of the corresponding travel trajectory data is low due to the relative position correction between the small regions. In such a case, accuracy can be improved by invalidating the travel trajectory data with low reliability and performing relative position correction again. Invalidating such travel trajectory data is performed in the first embodiment.

[0066] FIG. 14 also shows an example of a case where relative position correction within each small region is not performed correctly. In the example of FIG. 14, since relative position correction within each small region is not performed correctly, residual errors occur when relative position correction is performed between small regions. In this example, the relative position correction results within each small region appear correct at first glance, but residual errors occur in all of the driving trajectory data when relative position correction is performed between the small regions. In such a case, it can be determined that the reliability of relative position correction for the entire region is low. Accuracy can be improved by invalidating small regions with low reliability and performing relative position correction again. Invalidating such small regions will be performed in a second embodiment, which will be described later.

[0067] Next, the operation of the running locus generation device 14 according to the first embodiment will be described with reference to Fig. 15. Fig. 15 is a flowchart showing an example of the processing flow of the running locus generation program according to the first embodiment.

[0068] First, in step S100, the trajectory creation unit 142 generates travel trajectory data for each of the plurality of vehicles 12 using a known method based on various data stored in the data storage unit 140. Then, in step S100, the trajectory creation unit 142 stores the plurality of travel trajectory data in the travel trajectory data storage unit 144. The plurality of travel trajectory data stored in the travel trajectory data storage unit 144 is corrected by processing described below.

[0069] In step S102, the setting unit 146 reads out a plurality of pieces of travel locus data stored in the travel locus data storage unit 144. Then, in step S102, the setting unit 146 sets, for the plurality of pieces of travel locus data, a plurality of small areas that are located along the travel locus data and that have overlapping areas with adjacent small areas. As a result, for example, small areas such as those shown in FIG. 9(A) are set.

[0070] In step S104, the correction unit 148 corrects the traveling trajectory data captured in each of the plurality of small areas set in step S102, based on the surrounding targets captured in the plurality of traveling trajectory data. Specifically, as shown in Fig. 6 above, the correction unit 148 corrects the traveling trajectory data for each small area, based on the surrounding targets captured in the small area.

[0071] In step S106, for each pair of adjacent small areas set in step S102, the generation unit 150 corrects the position or tilt of at least one of the small areas in the pair so that the vehicle in the travel trajectory data that appears in the overlapping area of ​​the first small area of ​​the pair matches the vehicle in the travel trajectory data that appears in the overlapping area of ​​the second small area of ​​the pair. Specifically, the generation unit 150 corrects the position or tilt of the small areas as shown in Figure 9(C) above.

[0072] In step S108, generation unit 150 calculates, for each pair of small areas, the residual between the position of the vehicle captured in the overlapping area in the first area of ​​the pair and the position of the vehicle captured in the overlapping area in the second area of ​​the pair. Then, in step S108, generation unit 150 calculates a value that increases as the residual decreases, as a vehicle position reliability that indicates the reliability of the vehicle's position information, and assigns this value to the driving trajectory data.

[0073] In step S110, the generating unit 150 excludes (or invalidates) the traveling trajectory data whose vehicle position reliability is equal to or less than a predetermined threshold value from among the plurality of traveling trajectory data.

[0074] In step S112, the generation unit 150 determines whether a predetermined condition is satisfied. For example, the predetermined condition may be whether excluded (or invalidated) travel trajectory data exists, whether the processing of steps S100 to S110 has been repeated a predetermined number of times, or whether the sum of vehicle position reliabilities of a plurality of travel trajectory data is equal to or greater than a preset value. If the predetermined condition is satisfied, the process proceeds to step S114. On the other hand, if the predetermined condition is not satisfied, the process returns to step S104, and the processing of steps S104 to S110 is repeated.

[0075] In step S114, the generation unit 150 stores the plurality of pieces of travel locus data corrected in step S106 in the corrected travel locus data storage unit 152, and then ends the process.

[0076] As described above, the traveling trajectory generation device of the first embodiment sets, for the traveling trajectory data of each of the multiple vehicles, multiple small areas located along the traveling trajectory data and overlapping with adjacent small areas. The traveling trajectory generation device corrects the traveling trajectory data captured in each of the set multiple small areas based on surrounding objects captured in the traveling trajectory data of each of the multiple vehicles. Then, for each pair of adjacent small areas set in the traveling trajectory data of each of the multiple vehicles, the traveling trajectory generation device corrects the position or tilt of at least one of the pair of small areas so that the vehicle captured in the traveling trajectory data captured in the overlapping area of ​​the first small area of ​​the pair matches the vehicle captured in the traveling trajectory data captured in the overlapping area of ​​the second small area of ​​the pair, thereby correcting the traveling trajectory data of each of the multiple vehicles. This reduces errors that occur in the traveling trajectory data when dividing the traveling trajectory data of a moving object into multiple areas and correcting it. Specifically, when dividing multiple driving trajectory data into multiple regions and correcting the driving trajectory data, an overlapping area is set between the multiple regions, and alignment is performed within the overlapping area, thereby reducing the occurrence of positional deviations in the driving trajectory data between the multiple regions.

[0077] Furthermore, the traveling trajectory generation device of the first embodiment corrects the position or tilt of at least one of the pair of small areas, then calculates the residual between the position of the vehicle captured in the overlapping area in the first area and the position of the vehicle captured in the overlapping area in the second area, and calculates a value that increases as the residual decreases as the moving object position reliability, which indicates the reliability of the moving object position information.The traveling trajectory generation device then excludes, from the traveling trajectory data of each of the multiple vehicles, traveling trajectory data whose vehicle position reliability is equal to or less than a predetermined threshold.This makes it possible to obtain only traveling trajectory data with high reliability regarding the vehicle position.

[0078] Furthermore, the traveling trajectory generation device of the first embodiment generates traveling trajectory data for each of a plurality of vehicles by correcting the position or tilt of at least one of each pair of small areas, after excluding traveling trajectory data for which the vehicle position reliability is equal to or less than a predetermined threshold. This makes it possible to correct the plurality of traveling trajectory data again, after excluding traveling trajectory data for which the reliability of the vehicle position is low, and thereby generate traveling trajectory data with higher reliability.

[0079] [Second embodiment] Next, a second embodiment will be described. The second embodiment differs from the first embodiment in that after calculating the residual for each pair of small regions, the residual is used as the region reliability that represents the reliability of the pair of small regions.

[0080] The configurations of the running locus generation system 10 and the running locus generation device 14 of the second embodiment are the same as those of the first embodiment, and therefore detailed description thereof will be omitted.

[0081] As shown in Figures 11 and 14, the residuals calculated for each pair of sub-regions can also be assigned to that pair of sub-regions.

[0082] For this reason, as in the first embodiment, the generation unit 150 of the second embodiment corrects the position or inclination of at least one of the areas in each pair of small areas so that the vehicle in the driving trajectory data that is captured in the overlapping area in the first area of ​​the pair matches the vehicle in the driving trajectory data that is captured in the overlapping area in the second area of ​​the pair.

[0083] Next, the generation unit 150 of the second embodiment calculates the residual between the position of the vehicle reflected in the overlapping area in the first area and the position of the moving object reflected in the overlapping area in the second area, and calculates a value that increases as the residual decreases as the area reliability that represents the reliability of the pair of areas. Note that the smaller the residual, the higher the area reliability, and the larger the residual, the lower the area reliability.

[0084] Then, the generation unit 150 of the second embodiment excludes pairs of regions whose region reliability is equal to or less than a predetermined threshold value from the travel trajectory data of each of the plurality of vehicles.

[0085] Furthermore, the generation unit 150 of the second embodiment generates traveling trajectory data for each of the multiple vehicles by correcting the position or tilt of at least one of the small area pairs for each pair of small areas, after excluding (or invalidating) pairs of small areas whose area reliability is equal to or less than a predetermined threshold. This makes it possible to correct the multiple traveling trajectory data again, with low-reliability traveling trajectory data for the small areas excluded, and to generate more reliable traveling trajectory data.

[0086] Next, the operation of the running locus generation device 14 of the second embodiment will be described with reference to Fig. 16. Fig. 16 is a flowchart showing an example of the flow of processing in which the processing of steps S108 to S112 in the processing of the running locus generation program according to the first embodiment is changed.

[0087] The processes in steps S100 to S106 are executed in the same manner as in the first embodiment.

[0088] In step S208, the generation unit 150 of the second embodiment calculates the residual between the position of the vehicle reflected in the overlapping area in the first area and the position of the moving body reflected in the overlapping area in the second area, and calculates a value that increases as the residual decreases as the area reliability representing the reliability of the pair of areas.

[0089] In step S210, the generating unit 150 of the second embodiment excludes pairs of small regions whose region reliability is equal to or less than a predetermined threshold.

[0090] In step S212, the generation unit 150 of the second embodiment determines whether or not a predetermined condition is satisfied, as in the first embodiment. For example, the predetermined condition may be whether an excluded (or invalidated) small area exists, whether the processing of steps S100 to S210 has been repeated a predetermined number of times, or whether the sum of the area reliabilities of a plurality of pieces of traveling trajectory data is equal to or greater than a preset value. If the predetermined condition is satisfied, the process proceeds to step S114. On the other hand, if the predetermined condition is not satisfied, the process returns to step S104, and the processing of steps S104 to S210 is repeated.

[0091] As described above, the traveling trajectory generation device of the second embodiment corrects the position or tilt of at least one of the pair of small areas, then calculates the residual between the position of the vehicle captured in the overlapping area within the first area and the position of the vehicle captured in the overlapping area within the second area, and calculates a value that increases as the residual decreases as the area reliability, which represents the reliability of the pair of small areas.The traveling trajectory generation device then excludes, from the traveling trajectory data of each of the multiple vehicles, traveling trajectory data whose area reliability is equal to or less than a predetermined threshold.This makes it possible to obtain only small areas with high reliability regarding the vehicle's position.

[0092] Furthermore, the traveling trajectory generation device of the second embodiment generates traveling trajectory data for each of a plurality of vehicles by correcting the position or tilt of at least one of the small area pairs for each of the small area pairs after excluding traveling trajectory data for which the area reliability is equal to or less than a predetermined threshold. This makes it possible to correct the plurality of traveling trajectory data again after excluding small area pairs for which the reliability of the small areas is low, thereby generating traveling trajectory data with higher reliability.

[0093] The second embodiment can also be used in combination with the first embodiment.

[0094] [Other embodiments] The above describes an embodiment of a running locus generation device. The embodiment may be in the form of a program that causes a computer to function as each unit of the running locus generation device. The embodiment may be in the form of a computer-readable storage medium that stores the program.

[0095] Furthermore, the configuration of the running locus generation device described in the above embodiment is merely an example, and may be changed depending on the situation without departing from the spirit of the invention.

[0096] Furthermore, the processing flow of the program described in the above embodiment is merely an example, and unnecessary steps may be deleted, new steps may be added, or the processing order may be changed without departing from the scope of the invention. You can also swap the order.

[0097] In the above embodiment, the processing according to the embodiment is realized by a software configuration using a computer by executing a program, but the present invention is not limited to this. The embodiment may be realized by, for example, a hardware configuration or a combination of a hardware configuration and a software configuration.

[0098] Furthermore, the processes in the above-described embodiments may be stored as a program on a storage medium such as an optical disk and distributed.

[0099] In the above embodiment, the term "CPU" refers to a processor in a broad sense, including general-purpose processors and dedicated processors (e.g., GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, programmable logic device, etc.).

[0100] In this case, the operations of the processors may not only be performed by a single processor, but may also be performed by multiple processors located at physically separate locations working together. Furthermore, the order of the operations of the processors is not limited to the order described in the above embodiments, and may be changed as appropriate.

[0101] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference. [Example]

[0102] Fig. 17 is a diagram showing the results when actual travel locus data is corrected using each of the above-described embodiments. As shown in Fig. 17, by using each of the above-described embodiments, it is possible to correct the travel locus data without causing positional deviation of the travel locus data even at the boundaries of small regions.

[0103] (Implementation) The technology of the present disclosure may be implemented in the dependent relationships shown in the following appendices.

[0104] (Appendix 1) a setting unit that sets, for each of the travel locus data of a plurality of moving bodies, a plurality of regions that are located along the travel locus data and have overlapping regions with adjacent regions; a correction unit that corrects the travel locus data captured in each of the plurality of set areas based on the peripheral targets captured in the travel locus data of each of the plurality of moving bodies; a generating unit that generates the traveling trajectory data of each of the plurality of moving bodies by correcting the position or tilt of at least one of the adjacent regions set in the traveling trajectory data of each of the plurality of moving bodies so that the moving body in the traveling trajectory data that appears in the overlapping region in a first region of the pair coincides with the moving body in the traveling trajectory data that appears in the overlapping region in a second region of the pair; A driving trajectory generation device comprising: (Appendix 2) The generation unit for each pair of areas, correcting the position or tilt of at least one of the areas of the pair so that the moving object in the travel trajectory data that appears in the overlapping area in a first area of ​​the pair coincides with the moving object in the travel trajectory data that appears in the overlapping area in a second area of ​​the pair, then calculating a residual between the position of the moving object that appears in the overlapping area in the first area and the position of the moving object that appears in the overlapping area in the second area, and calculating a value that increases as the residual decreases as the value increases as the residual decreases as a moving object position reliability that indicates the reliability of the moving object position information; excluding, from the travel locus data of each of the plurality of moving bodies, travel locus data whose moving body position reliability is equal to or less than a predetermined threshold value; 2. The driving trajectory generation device according to claim 1. (Appendix 3) The generation unit After excluding the travel locus data of each of the plurality of moving bodies whose position reliability is equal to or less than a predetermined threshold value, correcting the position or tilt of at least one of the pairs of regions to correct the travel trajectory data of each of the plurality of moving bodies, thereby generating travel trajectory data of each of the plurality of moving bodies; 3. The driving trajectory generation device according to claim 2. (Appendix 4) The generation unit for each pair of areas, correcting the position or tilt of at least one of the areas of the pair so that a moving object in the travel trajectory data that appears in the overlapping area in a first area of ​​the pair coincides with a moving object in the travel trajectory data that appears in the overlapping area in a second area of ​​the pair, then calculating a residual between the position of the moving object that appears in the overlapping area in the first area and the position of the moving object that appears in the overlapping area in the second area, and calculating a value that increases as the residual decreases as the value increases as the residual decreases as a region reliability that represents a reliability for the region pair; excluding, from the travel trajectory data of each of the plurality of moving bodies, pairs of the regions whose region reliability is equal to or less than a predetermined threshold value; The running locus generation device according to any one of Supplementary notes 1 to 3. (Appendix 5) The generation unit After excluding the pair of regions whose region reliability is equal to or less than a predetermined threshold from the travel trajectory data of each of the plurality of moving bodies, correcting the position or tilt of at least one of the pairs of regions to correct the travel trajectory data of each of the plurality of moving bodies, thereby generating travel trajectory data of each of the plurality of moving bodies; 5. A driving trajectory generation device according to claim 4. (Appendix 6) For each of the travel locus data of a plurality of moving bodies, a plurality of regions are set that are located along the travel locus data and have adjacent and overlapping regions; correcting the travel locus data captured in each of the plurality of set areas based on the surrounding objects captured in the travel locus data of each of the plurality of moving bodies; for each pair of adjacent areas set in the travel trajectory data of each of the plurality of moving bodies, correcting the position or tilt of at least one area of ​​the pair so that the moving body in the travel trajectory data that appears in the overlapping area in a first area of ​​the pair coincides with the moving body in the travel trajectory data that appears in the overlapping area in a second area of ​​the pair, thereby generating the travel trajectory data of each of the plurality of moving bodies; A driving trajectory generation method in which processing is performed by a computer. (Appendix 7) Computer, a setting unit that sets, for each of the travel locus data of a plurality of moving bodies, a plurality of regions that are located along the travel locus data and have overlapping regions with adjacent regions; a correction unit that corrects the travel trajectory data captured in each of the plurality of set areas based on the peripheral targets captured in the travel trajectory data of each of the plurality of moving bodies; a generating unit that generates the travel trajectory data of each of the plurality of moving bodies by correcting the position or tilt of at least one of the adjacent regions set in the travel trajectory data of each of the plurality of moving bodies so that the moving body in the travel trajectory data that appears in the overlapping region in the first region of the pair coincides with the moving body in the travel trajectory data that appears in the overlapping region in the second region of the pair; A driving trajectory generation program to function as a [Explanation of symbols]

[0105] 10. Driving trajectory generation system 12 vehicles 14. Driving trajectory generation device 140 Data storage unit 142 Trajectory Creation Section 144 Travel trajectory data storage unit 146 Settings 148 Correction Unit 150 Generation part 152 Corrected driving trajectory data storage unit

Claims

1. a setting unit that sets, for each of the travel locus data of a plurality of moving bodies, a plurality of regions that are located along the travel locus data and have overlapping regions with adjacent regions; a correction unit that corrects the travel locus data captured in each of the plurality of set areas based on the peripheral targets captured in the travel locus data of each of the plurality of moving bodies; a generating unit that generates the traveling trajectory data of each of the plurality of moving bodies by correcting the position or tilt of at least one of the adjacent regions set in the traveling trajectory data of each of the plurality of moving bodies so that the moving body in the traveling trajectory data that appears in the overlapping region in a first region of the pair coincides with the moving body in the traveling trajectory data that appears in the overlapping region in a second region of the pair; A driving trajectory generation device comprising:

2. The generation unit correcting the position or tilt of at least one of the pairs of regions so that the moving body in the travel trajectory data that appears in the overlapping region in a first region of the pair coincides with the moving body in the travel trajectory data that appears in the overlapping region in a second region of the pair; then calculating a residual between the position of the moving body that appears in the overlapping region in the first region and the position of the moving body that appears in the overlapping region in the second region; and calculating a value that increases as the residual decreases as the value increases as the residual decreases as a moving body position reliability that indicates the reliability of the moving body's position information; excluding, from the travel locus data of each of the plurality of moving bodies, travel locus data whose moving body position reliability is equal to or less than a predetermined threshold value; The driving locus generating device according to claim 1 .

3. The generation unit After excluding the travel locus data of each of the plurality of moving bodies whose position reliability is equal to or less than a predetermined threshold value, correcting the position or tilt of at least one of the pairs of regions to correct the travel trajectory data of each of the plurality of moving bodies, thereby generating travel trajectory data of each of the plurality of moving bodies; The driving locus generating device according to claim 2 .

4. The generation unit for each pair of areas, correcting the position or tilt of at least one of the areas of the pair so that a moving body in the travel trajectory data that appears in the overlapping area in a first area of ​​the pair coincides with a moving body in the travel trajectory data that appears in the overlapping area in a second area of ​​the pair, then calculating a residual between the position of the moving body that appears in the overlapping area in the first area and the position of the moving body that appears in the overlapping area in the second area, and calculating a value that increases as the residual decreases as the value increases as the residual decreases as a region reliability that represents a reliability for the region pair; excluding, from the travel trajectory data of each of the plurality of moving bodies, pairs of the regions whose region reliability is equal to or less than a predetermined threshold value; The driving locus generating device according to claim 1 .

5. The generation unit After excluding the pair of regions whose region reliability is equal to or less than a predetermined threshold from the travel trajectory data of each of the plurality of moving bodies, correcting the position or tilt of at least one of the pairs of regions to correct the travel trajectory data of each of the plurality of moving bodies, thereby generating travel trajectory data of each of the plurality of moving bodies; The driving locus generating device according to claim 4.

6. For each of the travel locus data of a plurality of moving bodies, a plurality of regions are set that are located along the travel locus data and have adjacent and overlapping regions; correcting the travel locus data captured in each of the plurality of set areas based on the surrounding objects captured in the travel locus data of each of the plurality of moving bodies; for each pair of adjacent areas set in the travel trajectory data of each of the plurality of moving bodies, correcting the position or tilt of at least one area of ​​the pair so that the moving body in the travel trajectory data that appears in the overlapping area in a first area of ​​the pair coincides with the moving body in the travel trajectory data that appears in the overlapping area in a second area of ​​the pair, thereby generating the travel trajectory data of each of the plurality of moving bodies; A driving trajectory generation method in which processing is performed by a computer.

7. Computer, a setting unit that sets, for each of the travel locus data of a plurality of moving bodies, a plurality of regions that are located along the travel locus data and have overlapping regions with adjacent regions; a correction unit that corrects the travel trajectory data captured in each of the plurality of set areas based on the peripheral targets captured in the travel trajectory data of each of the plurality of moving bodies; a generating unit that generates the travel trajectory data of each of the plurality of moving bodies by correcting the position or tilt of at least one of the adjacent regions set in the travel trajectory data of each of the plurality of moving bodies so that the moving body in the travel trajectory data that appears in the overlapping region in the first region of the pair coincides with the moving body in the travel trajectory data that appears in the overlapping region in the second region of the pair; A driving trajectory generation program to function as a

Images