Position estimating system, map generating system, position estimating method, map generating method, and program

Abstract

Provided is a position estimating system comprising: a storage means for storing a point cloud map, which is a map in which position information and three-dimensional shapes of objects are associated with one other; an acquiring means for acquiring an image captured by a mobile body and point cloud data representing the three-dimensional shape of the object measured by the mobile body; a type determining means for determining the type of natural objects appearing in the acquired image; and a position estimating means for estimating the position of the mobile body by comparing the point cloud data, which excludes point cloud data corresponding to natural objects determined to be a type that changes over time, with the stored point cloud map.

Description

Position estimation system, map generation system, position estimation method, map generation method, and program

[0001] The present disclosure relates to a position estimation system, a map generation system, a position estimation method, a map generation method, and a program.

[0002] In Patent Document 1, the position of a moving object is estimated by matching tree trunks with an environmental map.

[0003] International Publication No. 2022 / 107588

[0004] However, when capturing images of natural objects such as trees to estimate the position of a moving object, there is a problem that the appearance of the natural objects changes depending on the season and time, making position estimation difficult. Patent Document 1 does not disclose how to deal with such changes in natural objects. Therefore, one of the objects of the present disclosure is to provide a position estimation system that is robust against changes in natural objects.

[0005] A position estimation system according to one aspect of the present disclosure includes: a storage means for storing a point cloud map, which is a map that associates position information with the three-dimensional shape of an object; an acquisition means for acquiring point cloud data representing images taken by a moving body and the three-dimensional shape of the object measured by the moving body; a type determination means for determining the type of natural object appearing in the acquired images; and a position estimation means for estimating the position of the moving body by comparing the acquired point cloud data, excluding point cloud data corresponding to the natural object determined to be a type that changes over time, with the stored point cloud map.

[0006] Another aspect of the position estimation system of the present disclosure is a position estimation system comprising: a storage means for storing a point cloud map, which is a map that associates position information with the three-dimensional shape of an object, and the types of natural objects included in the point cloud map; an acquisition means for acquiring point cloud data representing the three-dimensional shape of the object measured at the moving body; and a position estimation means for estimating the position of the moving body by comparing the acquired point cloud data with the stored point cloud map, excluding natural objects among the natural objects included in the point cloud map that are stored as types that change over time.

[0007] A map generation system according to one aspect of the present disclosure includes: an acquisition means for acquiring images captured by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and position information of the moving body; a type determination means for determining the type of natural object appearing in the acquired images; and a point cloud map generation means for generating a point cloud map, which is a map that associates position information with the three-dimensional shape of an object, based on the acquired position information and point cloud data excluding point cloud data corresponding to the natural object determined to be a type that changes over time from the acquired point cloud data.

[0008] Another aspect of the map generation system of the present disclosure is a map generation system comprising: an acquisition means for acquiring images taken by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and position information of the moving body; a type determination means for determining the type of natural object shown in the acquired images; a point cloud map generation means for generating a point cloud map, which is a map that associates position information with the three-dimensional shape of an object, based on the acquired point cloud data and the acquired position information; and a point cloud map update means for predicting the three-dimensional shape of a natural object determined to be of a type that changes over time, at a specified time point, after the change, and updating the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after the change.

[0009] A position estimation method according to one aspect of the present disclosure includes acquiring point cloud data representing images captured by a moving body and three-dimensional shapes of objects measured by the moving body, determining the types of natural objects appearing in the acquired images, and estimating the position of the moving body by comparing the acquired point cloud data, excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time, with a pre-stored point cloud map that associates position information with the three-dimensional shapes of objects.

[0010] Another aspect of the position estimation method of the present disclosure is a process of acquiring point cloud data representing a three-dimensional shape of an object measured on the moving body, and estimating the position of the moving body by comparing the acquired point cloud data with a pre-stored point cloud map that is a map that associates position information with the three-dimensional shape of the object, and estimating the position of the moving body by excluding natural objects that are pre-stored as being of a type that changes over time from the natural objects included in the point cloud map.

[0011] A map generation method according to one aspect of the present disclosure includes acquiring images captured by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and location information of the moving body, determining the type of natural object shown in the acquired images, and generating a point cloud map, which is a map that associates location information with the three-dimensional shape of the object, based on the acquired location information and point cloud data excluding point cloud data corresponding to the natural object determined to be a type that changes over time from the acquired point cloud data.

[0012] Another aspect of the map generation method of the present disclosure is a map generation method that acquires images taken by a moving body, point cloud data representing the three-dimensional shapes of objects measured by the moving body, and location information of the moving body, determines the type of natural object shown in the acquired images, generates a point cloud map that associates the location information with the three-dimensional shapes of the objects based on the acquired point cloud data and the acquired location information, predicts the three-dimensional shape of the natural object after change at a specified time point for the natural object determined to be of a type that changes over time, and updates the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after change.

[0013] A program according to one aspect of the present disclosure is a program that causes an information processing device to execute the following steps: acquire images taken by a moving body and point cloud data representing the three-dimensional shapes of objects measured by the moving body; determine the types of natural objects shown in the acquired images; and estimate the position of the moving body by comparing the acquired point cloud data, excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time, with a pre-stored point cloud map that associates position information with the three-dimensional shapes of objects.

[0014] Another aspect of the program of the present disclosure is a program that causes an information processing device to execute the following process: acquiring point cloud data representing the three-dimensional shape of an object measured on the moving body; and estimating the position of the moving body by comparing the acquired point cloud data with a pre-stored point cloud map that is a map that associates position information with the three-dimensional shape of the object, wherein the process estimates the position of the moving body by excluding natural objects that are pre-stored as being of a type that changes over time from the natural objects included in the point cloud map.

[0015] Another aspect of the program of the present disclosure is a program that causes an information processing device to execute the following steps: acquire images taken by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and location information of the moving body; determine the type of natural object shown in the acquired images; and generate a point cloud map that associates location information with the three-dimensional shape of the object based on the acquired location information and point cloud data excluding point cloud data corresponding to the natural object determined to be a type that changes over time from the acquired point cloud data.

[0016] Another aspect of the program of the present disclosure is a program that causes an information processing device to execute the following: acquire images taken by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and location information of the moving body; determine the type of natural object shown in the acquired images; generate a point cloud map that corresponds the location information to the three-dimensional shape of the object based on the acquired point cloud data and the acquired location information; predict the three-dimensional shape of the natural object after change at a specified time point for the natural object determined to be of a type that changes over time; and update the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after change.

[0017] The present disclosure provides a location estimation system that is robust against changes in natural objects.

[0018] FIG. 1 is a block diagram showing a configuration of a position estimation system according to the present disclosure. FIG. 2 is a block diagram showing a configuration of another position estimation system according to the present disclosure. FIG. 3 is a block diagram showing a configuration of another map generation system according to the present disclosure. FIG. 4 is a flowchart of a position estimation method according to the present disclosure. FIG. 5 is a flowchart of another position estimation method according to the present disclosure. FIG. 6 is a flowchart of a map generation method according to the present disclosure. FIG. 7 is a diagram showing an example of determining the type of natural object according to the present disclosure. FIG. 8 is a block diagram showing a configuration of an information processing device according to the present disclosure.

[0019] (Description of a Position Estimation System According to an Embodiment) FIG. 1 is a block diagram showing a configuration of a position estimation system according to the present disclosure. FIG. 9 is a diagram showing an example of determining the type of a natural object according to the present disclosure. Hereinafter, a configuration example of a position estimation system 100 will be described with reference to FIGS. 1 and 9. As shown in FIG. 1, the position estimation system 100 includes a storage unit 101, an acquisition unit 102, a type determination unit 103, and a position estimation unit 104. The position estimation system 100 is, for example, a system that captures an image of the surrounding environment and estimates a position based on the captured image.

[0020] The position estimation system 100 is attached to a mobile object and estimates the position of the mobile object. For example, the position estimation system 100 may be executed by an ECU (Electronic Control Unit). Alternatively, the position estimation system 100 may be provided remotely from the mobile object and estimate the position of the mobile object via electrical communication or the like. For example, the position estimation system 100 may be executed by an external cloud server. The mobile object can estimate its own position while moving using the position estimation system 100. An autonomous mobile object attached with a position estimation system is capable of autonomous driving.

[0021] The mobile body is, for example, a mobile body that moves outdoors with natural objects in the surrounding environment, such as a bus, automobile, or tractor. If there are natural objects indoors, the mobile body may also be a mobile body that moves indoors. The mobile body is, for example, an autonomous mobile body that is equipped with a prime mover or motor and can move autonomously using gasoline or electricity as power.

[0022] The storage unit 101 stores a point cloud map, which is a map that associates position information with the three-dimensional shape of an object. The point cloud map is created from point cloud data. Point cloud data is data that represents the three-dimensional shape of an object. The point cloud data may include color information in addition to the three-dimensional shape. The point cloud data is represented, for example, by a set of three-dimensional coordinates (X, Y, Z) and color information (R, G, B) of some points of an object.

[0023] The acquisition unit 102 acquires images captured by the mobile body and point cloud data representing the three-dimensional shape of an object measured by the mobile body. The mobile body is equipped with an imaging device. An RGB sensor is used as the imaging device. Alternatively, an infrared sensor may be used as the imaging device. The mobile body is also equipped with a distance sensor. The distance sensor includes a depth sensor, LiDAR (Light Detection and Ranging), laser, millimeter-wave radar, stereo camera, etc. The distance sensor is used to measure the distance to a natural object. The point cloud data is acquired using a distance sensor including, for example, LiDAR.

[0024] The type determination unit 103 determines the type of natural object shown in the acquired image. The natural object may be, for example, a tree or a plant. The type of natural object may be, for example, a deciduous tree, an evergreen tree, a perennial plant, or an annual plant. Natural objects change appearance over time, such as with the seasons. For example, evergreen trees have leaves all year round and their appearance changes little. However, deciduous trees sprout in spring, their leaves turn a deep green in summer, change color in autumn, and lose their leaves in winter. Therefore, the appearance of deciduous trees changes significantly throughout the year. Furthermore, perennial plants are plants that continue to bloom from the same stump in subsequent years. Therefore, the area of ​​perennial plants increases over time, significantly changing their appearance. Annual plants are plants that only grow for one year; they sprout in spring, grow in summer, and wither and disappear in autumn or winter. Therefore, annual plants significantly change their appearance over the course of a year.

[0025] 9, the position estimation system 100 detects deciduous trees, evergreen trees, and perennial grasses.

[0026] The image capturing device captures an image of the environment around the moving object and detects natural objects. The type determining unit 103 determines the type of the detected natural objects.

[0027] The position estimation unit 104 estimates the position of the moving object by comparing the acquired point cloud data, excluding point cloud data corresponding to natural objects determined to be of a type that changes over time, with a stored point cloud map. Specifically, the position estimation unit 104 compares a pre-created point cloud map with a current point cloud map to estimate the position of the moving object. The position estimation unit 104 excludes natural objects determined to be of a type that changes over time from the detected natural objects, and compares the current point cloud map with a pre-created point cloud map to estimate the position.

[0028] For example, in the case of deciduous trees, the leaves change throughout the year, so the map is created by excluding the point cloud of that part, and the position is estimated. For example, in the case of annual plants, the map is created by excluding the point cloud of that part, and the position is estimated. Therefore, the position is estimated using buildings, structures, buildings, walls, etc. other than natural objects.

[0029] This involves storing the classification of detected natural objects that are determined to be changing in type over time, and comparing the classification with the point cloud map.

[0030] For example, the tree classification is made as to whether it is an evergreen tree or a deciduous tree, and a map is created including the point cloud of that part, and this classification is taken into account when comparing it with the point cloud map. For example, if it is an evergreen tree, there is not much change in the point cloud map, so the currently acquired point cloud map is compared directly with a pre-created point cloud map. For example, if it is a deciduous tree, the point cloud map changes, so that part is excluded and the currently acquired point cloud map is compared with a pre-created point cloud map.

[0031] In addition, when the image contains a detected natural object that is determined to be of a changing type, the position estimation unit 104 may compare the acquired point cloud map with a point cloud map that has been updated to reflect a prediction of the changes over time of the natural object that changes over time.

[0032] For example, as described above, in summer, deciduous trees grow lush and become dark green. The position estimation unit 104 predicts seasonal and annual changes, updates a pre-created point cloud map, and compares it with a point cloud map acquired by a currently moving vehicle to estimate its position. For example, artificial intelligence (AI) is used for the prediction. For example, machine learning is performed to predict the appearance of lush leaves in advance, and when an image of a tree trunk is input, a point cloud map appropriate for the appearance of leaves on plants in that season is output.

[0033] With this configuration, a location estimation system is provided that determines the type of a natural object and estimates its location by excluding the natural object based on the determination or predicting changes in the natural object. Note that the storage unit 101, the acquisition unit 102, the type determination unit 103, and the location estimation unit 104 may be interpreted as a storage means, an acquisition means, a type determination means, and a location estimation means, respectively.

[0034] (Description of a Position Estimation System According to Another Embodiment) Fig. 2 is a block diagram showing the configuration of another position estimation system according to the present disclosure. An example of the configuration of a position estimation system 200 will be described below with reference to Fig. 2 .

[0035] 2 differs from the position estimation system 100 in that it does not have a type determination unit, that is, the types of natural objects included in the point cloud map are stored in advance.

[0036] As shown in FIG. 2 , the position estimation system 200 includes a storage unit 201, an acquisition unit 202, and a position estimation unit 203. The storage unit 201 stores a point cloud map, which is a map that associates position information with the three-dimensional shapes of objects, and the types of natural objects included in the point cloud map. The acquisition unit 202 acquires point cloud data that represents the three-dimensional shapes of objects measured on a mobile object. The position estimation unit 203 estimates the position of the mobile object by comparing the acquired point cloud data with the stored point cloud map. The position estimation unit 203 then performs the comparison while excluding natural objects included in the point cloud map that are stored as types that change over time.

[0037] The storage unit 201, the acquisition unit 202, and the position estimation unit 203 may be interpreted as a storage means, an acquisition means, a type determination means, and a position estimation means, respectively.

[0038] (Description of Map Generation System According to Embodiment) Fig. 3 is a block diagram showing the configuration of a map generation system according to the present disclosure. An example of the configuration of a map generation system 300 will be described below with reference to Fig. 3 .

[0039] As shown in FIG. 3 , the map generation system according to the present disclosure includes an acquisition unit 301 , a type determination unit 302 , and a point cloud map generation unit 303 .

[0040] The acquisition unit 301 acquires images captured by a moving object, point cloud data representing the three-dimensional shape of an object measured by the moving object, and position information of the moving object. The position information is acquired using, for example, a global positioning system (GPS).

[0041] The type determination unit 302 determines the type of natural object shown in the acquired image. As with the position estimation system 100, the type of natural object is determined using AI or the like.

[0042] The point cloud map generation unit 303 generates a point cloud map, which is a map that associates position information with the three-dimensional shape of an object, based on the acquired point cloud data, excluding point cloud data corresponding to natural objects determined to be of a type that changes over time, from the acquired point cloud data. The point cloud map generation unit 303 creates a point cloud map based on point cloud data acquired from a moving object. A point cloud map is a map that associates position information with the three-dimensional shape of an object. The point cloud map created by the point cloud map generation unit 303 is compared with a point cloud map created from point cloud data acquired in advance.

[0043] In this way, a map for the position estimation system is generated. The acquisition unit 301, the type determination unit 302, and the point cloud map generation unit 303 may be read as an acquisition means, a type determination means, and a point cloud map generation means, respectively.

[0044] (Description of a Map Generation System According to Another Embodiment) Fig. 4 is a block diagram showing the configuration of another map generation system according to the present disclosure. An example of the configuration of a map generation system 400 will be described below with reference to Fig. 4.

[0045] The map generation system 400 differs from the map generation system 300 in that it includes a point cloud map update unit 404. As shown in Fig. 4, the map generation system 400 includes an acquisition unit 401, a type determination unit 402, a point cloud map generation unit 403, and a point cloud map update unit 404.

[0046] The acquisition unit 401 acquires images captured by a moving object, point cloud data representing the three-dimensional shapes of objects measured by the moving object, and position information of the moving object. The type determination unit 402 determines the type of natural object captured in the acquired images.

[0047] The point cloud map generating unit 403 generates a point cloud map, which is a map that associates the position information with the three-dimensional shape of the object, based on the acquired point cloud data and the acquired position information.

[0048] The point cloud map update unit 404 predicts the three-dimensional shape of a natural object determined to be of a type that changes over time at a specified time point, and updates the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after the change.

[0049] In this way, a map can be generated that is used in a position estimation system that compares the updated point cloud map with the acquired point cloud map to estimate a position. The acquisition unit 401, the type determination unit 402, the point cloud map generation unit 403, and the point cloud map update unit 404 may be read as an acquisition means, a type determination means, a point cloud map generation means, and a point cloud map update means.

[0050] The position estimation system 100, the position estimation system 200, the map generation system 300, and the map generation system 400 are realized by an information processing device 1000 such as an ECU. FIG. 10 is a block diagram showing the configuration of the information processing device of the present disclosure. As shown in FIG. 10, the information processing device 1000 includes a processor 1001 that processes a program to execute processing, and a memory 1002 that stores the program. The information processing device 1000 may be configured as a single device or may be configured across multiple devices. The information processing device 1000 may also be configured as a cloud server that distributes some or all of its functions.

[0051] When the information processing device 1000 is configured as a single device, it can be said to be a position estimation device or a map generation device.

[0052] (Description of a position estimation method according to an embodiment) Fig. 5 is a flowchart of a position estimation method according to the present disclosure. The position estimation method will be described below with reference to the flowchart of Fig. 5. The position estimation method according to the present disclosure is executed by, for example, the position estimation system 100 of Fig. 1.

[0053] As shown in FIG. 5 , first, the acquisition unit 102 acquires images and point cloud data (step S501). The acquisition unit 102 acquires images captured by the mobile object and point cloud data representing the three-dimensional shapes of objects measured by the mobile object. Next, the type determination unit 103 determines the type (step S502). The type determination unit 103 determines the type of natural object captured in the acquired image. Next, the position estimation unit 104 estimates the position (step S503). The position estimation unit 104 estimates the position of the mobile object by comparing the point cloud data, from which point cloud data corresponding to natural objects determined to be of a type that changes over time, with a pre-stored point cloud map that associates position information with the three-dimensional shapes of objects. The position estimation unit 104 may estimate the position by excluding natural objects determined to be of a type that changes over time from among the detected natural objects. In addition, when the image contains a detected natural object that is determined to be of a changing type, the position estimation unit 104 may compare the point cloud map with an updated point cloud map that reflects a prediction of the changes over time of the natural object that changes over time.

[0054] (Description of a Location Estimation Method According to Another Embodiment) Fig. 6 is a flowchart of another location estimation method according to the present disclosure. Hereinafter, the another location estimation method will be described with reference to the flowchart of Fig. 6. The another location estimation method according to the present disclosure is executed by, for example, the location estimation system 200 of Fig. 2.

[0055] As shown in FIG. 6 , first, the acquisition unit 202 acquires point cloud data (step S601). The acquisition unit 202 acquires point cloud data representing the three-dimensional shape of an object measured on the mobile body. Next, the position estimation unit 203 estimates the position (step S602). The position estimation unit 203 estimates the position of the mobile body by comparing the acquired point cloud data with a pre-stored point cloud map that associates position information with the three-dimensional shape of the object. The position estimation unit 203 estimates the position of the mobile body by comparing the acquired point cloud data with the point cloud map, excluding natural objects that are pre-stored as being of a type that changes over time.

[0056] With this configuration, a position estimation method is provided that determines the type of natural object, and estimates the position by excluding the natural object or predicting changes in the natural object based on the determination.

[0057] (Description of Map Generation Method According to Embodiment) Fig. 7 is a flowchart of a map generation method according to the present disclosure. The map generation method according to the present disclosure will be described below using the flowchart of Fig. 7. The map generation method according to the present disclosure is executed, for example, by the map generation system 300 of Fig. 3. As shown in Fig. 7, first, the acquisition unit 301 acquires an image, point cloud data, and position information (step S701). The acquisition unit 301 acquires an image captured by a moving body, point cloud data representing a three-dimensional shape of an object measured by the moving body, and position information of the moving body.

[0058] Next, the type determination unit 302 determines the type (step S702). The type determination unit 302 determines the type of natural object shown in the acquired image. Next, the point cloud map generation unit 303 generates a point cloud map (step S703). The point cloud map generation unit 303 generates a point cloud map, which is a map that associates position information with the three-dimensional shape of an object, based on the acquired point cloud data excluding point cloud data corresponding to natural objects determined to be of a type that changes over time, and the acquired position information.

[0059] This configuration provides a map generation method for use in a location estimation system.

[0060] (Description of Another Map Generation Method According to an Embodiment) Fig. 8 is a flowchart of another map generation method according to the present disclosure. Hereinafter, the another map generation method according to the present disclosure will be described using the flowchart of Fig. 8. The another map generation method according to the present disclosure is executed, for example, by the map generation system 400 of Fig. 4. As shown in Fig. 8, first, the acquisition unit 401 acquires an image, point cloud data, and position information (step S801). The acquisition unit 401 acquires an image captured by a moving body, point cloud data representing a three-dimensional shape of an object measured by the moving body, and position information of the moving body.

[0061] Next, the type determination unit 402 determines the type (step S802). The type determination unit 402 determines the type of natural object captured in the acquired image. Next, the point cloud map generation unit 403 generates a point cloud map (step S803). Based on the acquired point cloud data and the acquired position information, the point cloud map generation unit 403 generates a point cloud map, which is a map that associates position information with the three-dimensional shape of an object. Next, the point cloud map update unit 404 updates the point cloud map (step S804). For natural objects determined to be of a type that changes over time, the point cloud map update unit 404 predicts the three-dimensional shape of the natural object after the change at a specified time point, and updates the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after the change.

[0062] This configuration provides a map generation method for use in a location estimation system.

[0063] Furthermore, some or all of the processes in the above-described position estimation system 100, position estimation system 200, map generation system 300, and map generation system 400 can be realized as a computer program. Such programs can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible recording media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, programmable ROMs (PROMs), erasable PROMs (EPROMs), flash ROMs, and random access memories (RAMs)). Furthermore, the programs may be supplied to a computer by various types of temporary computer-readable media. Examples of the temporary computer-readable medium include an electric signal, an optical signal, and an electromagnetic wave. The temporary computer-readable medium can provide the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

[0064] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0065] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0066] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes: (Supplementary Note 1) A position estimation system comprising: a storage means for storing a point cloud map, which is a map associating position information with the three-dimensional shape of an object; an acquisition means for acquiring point cloud data representing images taken by a moving body and the three-dimensional shape of the object measured by the moving body; a type determination means for determining the type of natural object shown in the acquired images; and a position estimation means for estimating the position of the moving body by collating point cloud data, excluding point cloud data corresponding to the natural object determined to be a type that changes over time, from the acquired point cloud data, with the stored point cloud map. (Supplementary Note 2) A position estimation system comprising: a storage means for storing a point cloud map, which is a map associating position information with the three-dimensional shapes of objects, and a type of natural object included in the point cloud map; an acquisition means for acquiring point cloud data representing the three-dimensional shape of the object measured by the mobile body; and a position estimation means for estimating the position of the mobile body by comparing the acquired point cloud data with the stored point cloud map, wherein the comparison excludes natural objects stored as a type that changes over time among the natural objects included in the point cloud map. (Supplementary Note 3) The position estimation system according to Supplementary Note 1 or 2, wherein the natural object is at least one of a tree and a grass. (Supplementary Note 4) The position estimation system according to any one of Supplements 1 to 3, wherein the type is at least one of an evergreen tree, a deciduous tree, a perennial herb, and an annual herb. (Supplementary Note 5) The position estimation system according to any one of Supplements 1 to 4, wherein the mobile body is a bus, an automobile, or a tractor. (Supplementary Note 6) A map generation system comprising: an acquisition means for acquiring images taken by a moving body, point cloud data representing three-dimensional shapes of objects measured by the moving body, and position information of the moving body; a type determination means for determining the type of natural object shown in the acquired images; and a point cloud map generation means for generating a point cloud map, which is a map that associates position information with the three-dimensional shapes of objects, based on the acquired position information and point cloud data excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time from the acquired point cloud data.(Supplementary Note 7) A map generation system comprising: an acquisition means for acquiring images captured by a mobile body, point cloud data representing three-dimensional shapes of objects measured by the mobile body, and position information of the mobile body; a type determination means for determining a type of natural object shown in the acquired images; a point cloud map generation means for generating a point cloud map that is a map correlating position information with the three-dimensional shapes of objects based on the acquired point cloud data and the acquired position information; and a point cloud map update means for predicting a three-dimensional shape after change at a specified time point for a natural object determined to be of a type that changes over time, and updating the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after change. (Supplementary Note 8) The map generation system according to Supplementary Note 6 or 7, wherein the natural object is at least one of a tree and a grass. (Supplementary Note 9) The map generation system according to any one of Supplementary Notes 6 to 8, wherein the type is at least one of an evergreen tree, a deciduous tree, a perennial herb, and an annual herb. (Supplementary Note 10) The map generation system according to any one of Supplementary Notes 6 to 9, wherein the moving body is a bus, an automobile, or a tractor. (Supplementary Note 11) A position estimation method comprising: acquiring images taken by a moving body and point cloud data representing three-dimensional shapes of objects measured by the moving body; determining the types of natural objects shown in the acquired images; and estimating the position of the moving body by collating the acquired point cloud data, excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time, with a pre-stored point cloud map that associates position information with the three-dimensional shapes of objects. (Supplementary Note 12) A position estimation method comprising: acquiring point cloud data representing a three-dimensional shape of an object measured on the moving body; and estimating a position of the moving body by comparing the acquired point cloud data with a pre-stored point cloud map, which is a map that associates position information with the three-dimensional shape of the object, and excluding natural objects that are pre-stored as being of a type that changes over time from the natural objects included in the point cloud map.(Supplementary Note 13) The position estimation method according to Supplementary Note 11 or 12, wherein the natural object is at least one of a tree and grass. (Supplementary Note 14) A map generation method comprising: acquiring images taken by a moving body, point cloud data representing three-dimensional shapes of objects measured by the moving body, and position information of the moving body, determining the types of natural objects shown in the acquired images, and generating a point cloud map that associates position information with the three-dimensional shapes of objects based on the acquired position information and point cloud data excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time from the acquired point cloud data. (Supplementary Note 15) A map generation method comprising: acquiring images captured by a moving body, point cloud data representing three-dimensional shapes of objects measured by the moving body, and position information of the moving body; determining the type of natural object shown in the acquired images; generating a point cloud map that associates position information with the three-dimensional shapes of objects based on the acquired point cloud data and the acquired position information; predicting the three-dimensional shape of the natural object after change at a specified time point for the natural object determined to be of a type that changes over time; and updating the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after change. (Supplementary Note 16) A program that causes an information processing device to execute the following steps: acquire images taken by a moving body and point cloud data representing the three-dimensional shapes of objects measured by the moving body; determine the types of natural objects shown in the acquired images; and estimate the position of the moving body by comparing the acquired point cloud data, excluding point cloud data corresponding to the natural objects determined to be of a type that changes over time, with a pre-stored point cloud map that associates position information with the three-dimensional shapes of objects.(Supplementary Note 17) A program causing an information processing device to execute the following steps: acquiring point cloud data representing a three-dimensional shape of an object measured by the moving body; and estimating a position of the moving body by comparing the acquired point cloud data with a pre-stored point cloud map that is a map that associates position information with the three-dimensional shape of the object, wherein the processing estimates the position of the moving body by excluding natural objects that are pre-stored as being of a type that changes over time from the natural objects included in the point cloud map. (Supplementary Note 18) The program according to Supplementary Note 16 or 17, wherein the natural objects are at least one of trees and grass. (Supplementary Note 19) A program that causes an information processing device to execute the following steps: acquire images taken by a moving body, point cloud data representing the three-dimensional shape of an object measured by the moving body, and position information of the moving body; determine the type of natural object shown in the acquired images; and generate a point cloud map that associates position information with the three-dimensional shape of an object based on the acquired position information and point cloud data excluding point cloud data corresponding to the natural object determined to be a type that changes over time from the acquired point cloud data. (Supplementary Note 20) A program that causes an information processing device to execute the following steps: acquire images taken by a moving body, point cloud data representing the three-dimensional shapes of objects measured by the moving body, and location information of the moving body; determine the type of natural object shown in the acquired images; generate a point cloud map that associates location information with the three-dimensional shapes of objects based on the acquired point cloud data and the acquired location information; predict the three-dimensional shape of the natural object after change at a specified time point that is determined to be of a type that changes over time; and update the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after change.

[0067] Some or all of the elements (e.g., configurations and functions) described in Appendix 3 to Appendix 5 that are dependent on Appendix 1 and Appendix 2 (e.g., system) may also be dependent on Appendix 11 and Appendix 12 (e.g., method) and Appendix 16 and Appendix 17 (e.g., program) in the same dependency relationship as Appendix 3 to Appendix 5. Also, some or all of the elements (e.g., configurations and functions) described in Appendix 8 to Appendix 10 that are dependent on Appendix 6 and Appendix 7 (e.g., system) may also be dependent on Appendix 14 and Appendix 15 (e.g., method) and Appendix 19 and Appendix 20 (e.g., program) in the same dependency relationship as Appendix 8 to Appendix 10. Some or all of the elements described in any appendix may be applicable to various hardware, software, recording means for recording software, systems, and methods.

[0068] 100 Position estimation system, 101 Memory unit, 102 Acquisition unit, 103 Type determination unit, 104 Position estimation unit, 200 Position estimation system, 201 Memory unit, 202 Acquisition unit, 203 Position estimation unit, 300 Map generation system, 301 Acquisition unit, 302 Type determination unit, 303 Point cloud map generation unit, 400 Map generation system, 401 Acquisition unit, 402 Type determination unit, 403 Point cloud map generation unit, 404 Point cloud map update unit, 1000 Information processing device, 1001 Processor, 1002 Memory

Claims

1. A storage means for storing a point cloud map, which is a map that associates location information with the three-dimensional shape of an object, Acquisition means for acquiring images captured by a moving object and point cloud data representing the three-dimensional shape of an object measured by the moving object, A type determination means for determining the type of natural object shown in the acquired image, A position estimation means for estimating the position of the moving object by comparing the point cloud data obtained by excluding point cloud data corresponding to natural objects that are determined to be of a type that changes over time with the stored point cloud map, A position estimation system equipped with the following features.

2. A point cloud map, which is a map that associates location information with the three-dimensional shape of an object, and a storage means for storing the types of natural objects included in the point cloud map, Acquisition means for acquiring point cloud data representing the three-dimensional shape of an object measured by the moving body, A position estimation means for estimating the position of a moving object by comparing the acquired point cloud data with the stored point cloud map, the position estimation means for excluding natural objects stored as being of a type that changes over time from among the natural objects included in the point cloud map and comparing them, A position estimation system equipped with the following features.

3. Acquisition means for acquiring images captured by a moving object, point cloud data representing the three-dimensional shape of an object measured by the moving object, and position information of the moving object, A type determination means for determining the type of natural object shown in the acquired image, A point cloud map generation means generates a point cloud map, which is a map that associates the location information with the three-dimensional shape of an object, based on the point cloud data obtained by excluding point cloud data corresponding to the natural object that is determined to be of a type that changes over time from the acquired point cloud data, and the acquired location information. A map generation system equipped with the following features.

4. Acquisition means for acquiring images captured by a moving object, point cloud data representing the three-dimensional shape of an object measured by the moving object, and position information of the moving object, A type determination means for determining the type of natural object shown in the acquired image, A point cloud map generation means generates a point cloud map, which is a map that associates the location information with the three-dimensional shape of an object, based on the acquired point cloud data and the acquired location information. A point cloud map updating means predicts the three-dimensional shape of a natural object that has been determined to be of a type that changes over time at a specified point in time, and updates the three-dimensional shape of the natural object included in the point cloud map with the predicted three-dimensional shape after the change. A map generation system equipped with the following features.

5. Images captured by the moving object and point cloud data representing the three-dimensional shape of the object measured by the moving object are acquired. Determine the type of natural object shown in the acquired image, The position of the moving object is estimated by comparing the point cloud data obtained, which excludes point cloud data corresponding to natural objects that are determined to be of a type that changes over time, with a point cloud map that is pre-stored and is a map that associates positional information with the three-dimensional shape of an object. Location estimation method.

6. Point cloud data representing the three-dimensional shape of the object measured by the moving body is acquired. A process for estimating the position of a moving object by comparing the acquired point cloud data with a point cloud map that is a map which associates positional information with the three-dimensional shape of an object, wherein the process for estimating the position of the moving object is performed by excluding natural objects that are pre-stored as being of a type that changes over time from among the natural objects included in the point cloud map and comparing them. Location estimation method.

7. The mobile device acquires images captured by the mobile device, point cloud data representing the three-dimensional shape of an object measured by the mobile device, and position information of the mobile device. Determine the type of natural object shown in the acquired image, Based on the point cloud data obtained by excluding point cloud data corresponding to natural objects that are determined to be of a type that changes over time, and the acquired position information, a point cloud map is generated, which is a map that associates position information with the three-dimensional shape of an object. Map generation method.

8. The mobile device acquires images captured by the mobile device, point cloud data representing the three-dimensional shape of an object measured by the mobile device, and position information of the mobile device. Determine the type of natural object shown in the acquired image, Based on the acquired point cloud data and the acquired position information, a point cloud map is generated, which is a map that associates the position information with the three-dimensional shape of the object. For natural objects determined to be of a type that changes over time, the three-dimensional shape after the change at a specified point in time is predicted, and the three-dimensional shape of the natural object included in the point cloud map is updated with the predicted three-dimensional shape after the change. Map generation method.

9. Images captured by the moving object and point cloud data representing the three-dimensional shape of the object measured by the moving object are acquired. Determine the type of natural object shown in the acquired image, The position of the moving object is estimated by comparing the point cloud data obtained, which excludes point cloud data corresponding to natural objects that are determined to be of a type that changes over time, with a point cloud map that is pre-stored and is a map that associates positional information with the three-dimensional shape of an object. A program that instructs an information processing device to perform a specific action.

10. Point cloud data representing the three-dimensional shape of the object measured by the moving body is acquired. A process for estimating the position of a moving object by comparing the acquired point cloud data with a pre-stored point cloud map, which is a map that associates positional information with the three-dimensional shape of an object, wherein the process for estimating the position of the moving object is performed by excluding natural objects that are pre-stored as being of a type that changes over time from among the natural objects included in the point cloud map and comparing them. A program that instructs an information processing device to perform a specific action.