Server, information processing device, server program, information processing program, and position estimation system

Abstract

To provide a server capable of improving self-position estimation accuracy as compared with a case where a vehicle estimates its own position independently.SOLUTION: A server 3 is configured to: estimate, for each time, a cyber-space position of a vehicle by using edge data for each time including the estimated real-space position of a vehicle P and information related to an environment in which the vehicle travels; correct the estimated cyber-space position of the vehicle and the time at which the vehicle is present at the estimated cyber-space position on the basis of an error between the vehicle's real-space position and the vehicle's cyber-space position, and an error between an edge time and a system time; and transmit information including the corrected cyber-space position of the vehicle to an information processing device 2 which has estimated the vehicle's real-space position.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a server, an information processing device, a server program, an information processing program, and a position estimation system. [Background technology]

[0002] Patent Document 1 discloses an information processing method in which a first information processing device sets a first coordinate axis in a specified space, a second information processing device sets a second coordinate axis in the specified space, the second information processing device generates a conversion rule for converting the second coordinate axis into the first coordinate axis, and each of the first and second information processing devices acquires a point cloud represented by the first coordinate axis from an image captured of the specified space, and creates an environmental map using the acquired point cloud.

[0003] Patent Document 2 discloses a surveying system for acquiring three-dimensional point cloud data that includes an information processing device and a surveying device, and that includes a scanner unit that performs measurements to acquire the three-dimensional point cloud data of the surveying device, a measurement position designation unit that designates the measurement position where the three-dimensional point cloud data should be acquired, a position acquisition unit that acquires the self-position of the surveying device, and a measurement calculation unit that performs calculations to measure the measurement position from the self-position. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-34234 [Patent Document 2] Japanese Patent Publication No. 2022-145442 Summary of the Invention [Problem to be solved by the invention]

[0005] In recent years, technologies that use map data to estimate a vehicle's current location, i.e., its own position, have been used in a variety of fields. In particular, in the field of transportation, in order to realize automated driving of vehicles, it is necessary to estimate the vehicle's position on a map as accurately as possible.

[0006] When such vehicles estimate their own position, they use measurements from various sensors, such as LiDAR (Light Detection and Ranging) that captures distance images, and acceleration sensors. However, because other vehicles are also traveling on the road, other vehicles, for example, can act as obstacles, making it possible to obtain only a partial view of the shape of the road and buildings around the vehicle, which can reduce the accuracy of estimating their own position.

[0007] Furthermore, for example, when a vehicle is traveling on a straight road with no nearby landmarks, or on a mountain road with a continuous, identical landscape, there may be multiple locations in the map data that are similar to the environment around the vehicle obtained from various sensors, and the vehicle's location may be estimated to be a location different from its actual location.

[0008] The present disclosure has been made in consideration of the above circumstances, and aims to provide a server, an information processing device, a server program, an information processing program, and a position estimation system that can improve the accuracy of estimating a vehicle's own position compared to when the vehicle estimates its own position alone. [Means for solving the problem]

[0009] The server (3) of the present disclosure includes an estimation unit (3A) that estimates the position of a vehicle (P) at each time in cyberspace using edge data at each time including information about the position of the vehicle estimated in real space and the environment in which the vehicle is traveling; a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on the errors between the position of the vehicle in real space and the time in cyberspace; and a communication unit (3C) that transmits information about the corrected position of the vehicle corrected by the correction unit to the information processing device (2) that estimated the position of the vehicle in real space.

[0010] The information processing device (2) of the present disclosure includes an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that represents the degree of agreement between the environment on the map data at the vehicle's position estimated by the edge estimation unit and the surrounding environment recognized by the vehicle, and associates the calculated edge score with the vehicle's position at each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits the vehicle's position estimated at each time and information regarding the surrounding environment at each vehicle position, including the edge score and the measurement values ​​of the sensor, as edge data to a server (3) configured in cyberspace that has an environment identical to real space; and an edge correction unit (2D) that corrects the vehicle's position estimated by the edge estimation unit using information regarding the vehicle's position in cyberspace received from the server by the edge communication unit.

[0011] The server program (21) of the present disclosure is a program for causing a computer to execute a process of estimating the position of a vehicle (P) at each time in cyberspace using edge data at each time including information about the position of the vehicle estimated in real space and the environment in which the vehicle is traveling, correcting the estimated position of the vehicle in cyberspace and the time at which the vehicle is present at the estimated position based on the errors between the position of the vehicle in real space and the time in cyberspace, and transmitting information about the corrected position of the vehicle to the information processing device (2) that estimated the position of the vehicle in real space.

[0012] The information processing program (11) disclosed herein is a program for causing a computer to estimate the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4), calculate an edge score that represents the degree of correspondence between the environment on the map data at the estimated position of the vehicle and the surrounding environment recognized by the vehicle, associate the calculated edge score with the position of the vehicle at each estimated time, transmit the estimated position of the vehicle at each time and information on the surrounding environment at each position of the vehicle, including the edge score and the measurement values ​​of the sensor, as edge data to a server (3) configured in cyberspace that has an environment identical to real space, and correct the estimated position of the vehicle using information on the position of the vehicle in cyberspace received from the server.

[0013] The position estimation system (1) of the present disclosure includes an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that indicates the degree of agreement between an environment on the map data at the position of the vehicle estimated by the edge estimation unit and a surrounding environment recognized by the vehicle, and associates the calculated edge score with the position of the vehicle at each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits the position of the vehicle estimated at each time and information about the surrounding environment at each position of the vehicle, including the edge score and the measurement values ​​of the sensor, as edge data to a server (3) configured in cyberspace that represents an environment identical to the real space; and an edge communication unit (2C) that transmits the edge data received from the server by the edge communication unit. The server (3) includes an information processing device (2) having an edge correction unit (2D) that corrects the position of the vehicle estimated by the edge estimation unit using information about the position of the vehicle in space, and an estimation unit (3A) that estimates the position of the vehicle at each time in cyberspace using the edge data for each time including the position of the vehicle estimated in real space and information about the environment in which the vehicle is traveling, received from the information processing device, a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on errors between the position of the vehicle in real space and the time in cyberspace, and a communication unit (3C) that transmits information about the position of the vehicle corrected by the correction unit to the information processing device (2). [Effects of the Invention]

[0014] According to the present disclosure, it is possible to obtain an effect that the accuracy of estimating the vehicle's own position can be improved compared to when the vehicle estimates its own position independently. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a position estimation system. [Figure 2] FIG. 10 is a diagram illustrating an example of recognition data. [Figure 3] FIG. 2 is a diagram illustrating an example of a functional configuration of an information processing device. [Figure 4] FIG. 10 is a diagram showing an example of a vehicle position estimated using map data. [Figure 5] FIG. 2 illustrates an example of a functional configuration of a server. [Figure 6] FIG. 1 is a diagram illustrating an example of an environment in cyberspace in which a vehicle travels at each time. [Figure 7] FIG. 10 is a diagram illustrating an example of a matching process between an environment in cyberspace and a surrounding environment represented by recognition data. [Figure 8] FIG. 10 is a diagram showing an example of estimation of a cyberspace candidate position of a vehicle at each time. [Figure 9] FIG. 10 is a diagram showing an example of an estimated position of a vehicle estimated using the amount of movement of the vehicle. [Figure 10] FIG. 10 is a diagram showing an example in which cyberspace candidate positions of a vehicle are classified into groups along a time series. [Figure 11] FIG. 10 is a diagram illustrating an example of correcting the cyberspace position of a vehicle. [Figure 12] FIG. 1 is a diagram illustrating an example of the configuration of an information processing device implemented by a computer. [Figure 13] FIG. 10 is a diagram illustrating an example of the configuration of a server configured by a computer. [Figure 14] 10 is a flowchart illustrating an example of the flow of an estimation process executed by an information processing device. [Figure 15] 10 is a flowchart illustrating an example of the flow of edge data accumulation processing executed by the server. [Figure 16] 10 is a flowchart illustrating an example of the flow of an estimation process executed by the server. [Figure 17] 10 is a flowchart showing an example of the flow of a candidate position estimation process. [Figure 18] 10 is a flowchart showing an example of the flow of a position estimation process. [Figure 19]10 is a flowchart illustrating an example of the flow of a correction process. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, the present embodiment will be described with reference to the drawings. The same components and processes are denoted by the same reference numerals throughout the drawings, and duplicated explanations will be omitted. The dimensional proportions in the drawings are exaggerated for the sake of explanation, and may differ from the actual proportions.

[0017] FIG. 1 is a diagram showing an example of the configuration of a position estimation system 1 that estimates the position of a vehicle traveling on a road 6A.

[0018] Among multiple vehicles, the vehicle of interest whose position is to be estimated will be referred to as "vehicle P," and the vehicles traveling around vehicle P will be referred to as "vehicle Q." When there is no need to distinguish between vehicle P and vehicle Q, they will simply be referred to as "vehicles."

[0019] The vehicle P is equipped with an information processing device 2, which performs wireless communication with a wireless communication device 7 provided in the vicinity of the road 6A.

[0020] Each wireless communication device 7 is connected to a communication network 5, and the server 3 is also connected to the communication network 5. Therefore, the information processing device 2 of the vehicle P exchanges various information with the server 3 through the wireless communication device 7.

[0021] As will be described later, for example, the information processing device 2 estimates the position of the vehicle P at each time using the measurement values ​​of various sensors 17 (see FIG. 12) attached to the vehicle P and the map data 4 (see FIG. 3). Then, the information processing device 2 transmits to the server 3 information on the environment in which the vehicle P is traveling, such as the measurement values ​​of the various sensors 17 attached to the vehicle P and road conditions estimated from the measurement values ​​of the various sensors 17, as well as the estimated position of the vehicle P at each time.

[0022] Fig. 2 is a diagram showing an example of road conditions estimated from measurement values ​​of various sensors 17. The surrounding road conditions recognized by vehicle P as shown in Fig. 2 can be obtained from the measurement values ​​of various sensors 17, and therefore a data set representing road conditions in real space reproduced from the measurement values ​​of various sensors 17 attached to vehicle P is referred to as "recognition data 8." The recognition data 8 is an example of information about the environment in which vehicle P is traveling.

[0023] As shown in FIG. 1, a collection device 9 is installed around the road 6A. The collection device 9 is equipped with various sensors (not shown), such as cameras and beacons, and collects data representing road conditions that change over time, such as the position of vehicles on the road 6A and the level of congestion. The data representing road conditions collected by the collection device 9 is called "surrounding data" because it represents the conditions around the road 6A on which the vehicle P is traveling. The surrounding data is also an example of information about the environment in which the vehicle P is traveling.

[0024] Each collection device 9 is connected to the communication network 5 in the same way as the wireless communication device 7. Each collection device 9 transmits the collected peripheral data to the server 3 via the communication network 5.

[0025] The server 3 receives information obtained in the real space, uses the received information to build a virtual environment that represents the same environment as the real space in the memory of the server 3, and executes various simulations using the virtual environment. Hereinafter, the virtual space in the memory where the virtual environment is built will be referred to as "cyberspace."

[0026] Specifically, the server 3 uses the surrounding data received from the collection device 9 to construct a virtual environment in cyberspace around the road 6A on which the vehicle P is traveling, and estimates the position of the vehicle P in the virtual environment at each time point using the position information and recognition data 8 received from the vehicle P. In other words, the position estimation system 1 estimates the position of the vehicle P using a digital twin.

[0027] The server 3 transmits information relating to the position of the vehicle P estimated in cyberspace to the information processing device 2, and the information processing device 2 corrects the position of the vehicle P estimated by the information processing device 2 using the information received from the server 3. Note that the method for estimating the position of the vehicle P in the information processing device 2 and the server 3 will be described in detail later.

[0028] For convenience of explanation, information obtained in real space will be collectively referred to as “edge data,” and information obtained in cyberspace will be collectively referred to as “cyber data.” Furthermore, the information processing device 2 is not shown in the vehicle P illustrated in Figures 2 and subsequent drawings.

[0029] Fig. 3 is a diagram illustrating an example of the functional configuration of the information processing device 2. As illustrated in Fig. 3, the information processing device 2 includes functional units, namely, an edge estimation unit 2A, an edge calculation unit 2B, an edge communication unit 2C, an edge correction unit 2D, and an edge storage unit 2E, as well as an edge time 30A.

[0030] The edge estimation unit 2A estimates the position of the vehicle P in real space at each time using the map data 4 and the measurement values ​​of the various sensors 17 obtained at each time.

[0031] The map data 4 is data representing the spatial environment in which the vehicle P travels, and is three-dimensional map information including structural information of the road 6A, such as the number and width of lanes, supplementary information such as the positions of traffic lights and the positions and contents of signs, and the positions and shapes of buildings around the road 6A. The map data 4 is stored, for example, in the edge storage unit 2E.

[0032] The edge estimation unit 2A generates recognition data 8 using measurement values ​​of various sensors 17, and estimates the position on the map data 4 that is most similar to the environment represented by the recognition data 8 as the position of the vehicle P.

[0033] The position of vehicle P is estimated at predetermined intervals, for example, every 5 seconds. The interval at which the edge estimation unit 2A estimates the position of vehicle P is called the "unit time Δt." The unit time Δt can be changed by the user. Each time at which the position of vehicle P is estimated is also called a "frame."

[0034] The estimated position of vehicle P is associated with the time at which vehicle P is thought to have been present at the estimated position. The time associated with the estimated position of vehicle P is referred to as "edge time 30A." Edge time 30A is set based on the time of information processing device 2. For example, if edge estimation unit 2A estimates the position of vehicle P based on recognition data 8 generated using measurement values ​​of various sensors 17 acquired at 15:30:30 on March 21, 2024, the estimated position of vehicle P will be the position of vehicle P at 15:30:30 on March 21, 2024.

[0035] In the edge estimation unit 2A, a known matching process is used as a method for detecting a position similar to the environment represented by the recognition data 8 from the spatial environment in which the vehicle P travels, which is obtained from the map data 4. FIG. 4 is a diagram showing an example of the position of the vehicle P estimated by the matching process between the map data 4 and the recognition data 8.

[0036] If vehicle Q is traveling around vehicle P, vehicle Q acts as an obstacle, making it possible to acquire only a partial image of the real-space environment in which vehicle P is traveling, which may reduce the accuracy of estimating the position of vehicle P. Furthermore, similar environments occur on straight roads with no landmark buildings or mountain roads with the same scenery continuing, and therefore multiple positions on map data 4 that are similar to the environment represented by recognition data 8 may be detected, which may reduce the accuracy of estimating vehicle P's position.

[0037] Therefore, the edge calculation unit 2B calculates an edge score that represents the degree of agreement between the environment on the map data 4 at the position of the vehicle P estimated by the edge estimation unit 2A and the surrounding environment recognized by the vehicle P, i.e., the recognition data 8. For example, the larger the edge score value, the more similar the respective environments are. In other words, the edge score is a value that represents the accuracy of the position of the vehicle P estimated by the edge estimation unit 2A. As an example, the minimum value of the edge score is set to "0".

[0038] The edge calculation unit 2B associates the calculated edge scores with the positions of the vehicle P at each time estimated by the edge estimation unit 2A.

[0039] The edge communication unit 2C transmits to the server 3 as edge data, for example, the position of vehicle P estimated by the edge estimation unit 2A at each time, the edge score associated with the estimated position of vehicle P at each time, the measurement values ​​of various sensors 17 at each time, and information regarding the surrounding environment at each position of vehicle P, including recognition data 8 at each time.

[0040] As can be seen from the fact that the edge score represents the reliability of the position of vehicle P estimated by edge estimation unit 2A, the position of vehicle P estimated by edge estimation unit 2A does not necessarily represent the correct position. Therefore, edge correction unit 2D corrects the position of vehicle P at each time estimated by edge estimation unit 2A using information regarding the position of vehicle P in cyberspace that edge communication unit 2C receives from server 3 in response to the transmitted edge data.

[0041] The information processing device 2 does not necessarily have to be installed in the vehicle P, and may be installed, for example, inside a building. In this case, the measurement values ​​of the various sensors 17 attached to the vehicle P are transmitted to the information processing device 2 through the communication network 5.

[0042] The time set in the collection device 9 is also associated with the peripheral data collected by the collection device 9 as the edge time 30A.

[0043] On the other hand, FIG. 5 is a diagram illustrating an example of the functional configuration of the server 3 that receives edge data from the information processing device 2.

[0044] As shown in FIG. 5, the server 3 includes functional units, ie, an estimation unit 3A, a correction unit 3B, and a communication unit 3C, as well as a system time 30B.

[0045] The estimation unit 3A estimates the position of the vehicle P at each time in cyberspace using the position of the vehicle P in real space estimated by the information processing device 2 and edge data at each time including information about the environment in which the vehicle P is traveling.

[0046] As already explained, the information about the environment in which the vehicle P is traveling includes, for example, measurement values ​​of various sensors 17 attached to the vehicle P, the recognition data 8, and surrounding data transmitted from the collection device 9. The estimation unit 3A uses the surrounding data included in the edge data and the position of the vehicle P in real space to construct in cyberspace the environment in which the vehicle P is traveling at each time.

[0047] Fig. 6 is a diagram showing an example of the environment in which vehicle P travels at each time. The example shown in Fig. 6 shows the environment in which vehicle P travels at time t, the environment in which vehicle P travels at time t-1, which is the frame immediately before time t, and the environment in which vehicle P travels at time t+1, which is the frame immediately after time t. Specifically, the example shown in Fig. 6 shows how the environment of road 6B on which vehicle P travels is constructed for each time using surrounding data that represents the environment within a predetermined range including the position of vehicle P estimated in real space.

[0048] If n is an integer equal to or greater than 0, time t+n represents the time nΔt ahead of time t, and time tn represents the time nΔt before time t. Note that times t±n represent the times at edge time 30A.

[0049] For ease of explanation, the position of vehicle P estimated in real space may be referred to as the "real space position of vehicle P," and the position of vehicle P in cyberspace may be referred to as the "cyber space position of vehicle P." Also, to avoid confusion, road 6A is a road in real space, while road 6B is a road in cyberspace.

[0050] The estimation unit 3A executes a matching process to detect similar positions while moving the surrounding environment at each time of the position of the vehicle P estimated in real space by the information processing device 2, which is the environment represented by the recognition data 8, over the entire environment of the road 6B constructed from the surrounding data at the same time. There are no restrictions on the range of the environment of the road 6B constructed from the surrounding data, and it can be set in advance by the user, for example.

[0051] Hereinafter, the environment constructed in cyberspace using surrounding data will be referred to as the "environment in cyberspace," and the environment represented by the recognition data 8, which is the surrounding environment at the position of vehicle P estimated in real space by the information processing device 2, will be referred to as the "surrounding environment represented by the recognition data 8."

[0052] 7 is a diagram showing an example of a matching process between an environment in cyberspace at a specific time and a surrounding environment represented by recognition data 8. For example, the estimation unit 3A moves the surrounding environment represented by recognition data 8 surrounded by a dotted line 28 throughout the entire environment in cyberspace along the direction of arrow U, and calculates the cyber score at each destination position. In FIG. 7, shaded vehicles indicate vehicles in the surrounding environment represented by recognition data 8, and unshaded vehicles indicate vehicles in the environment in cyberspace.

[0053] The cyber score is a value that represents the degree of correspondence between the environment in cyberspace at each time and the surrounding environment represented by the recognition data 8. Similar to the edge score, for example, the larger the cyber score, the more similar the environments are. In other words, the cyber score is a value that represents the accuracy of the position of vehicle P estimated by the estimation unit 3A. As an example, the minimum cyber score is set to "0."

[0054] In estimating the cyberspace position of the vehicle P by the estimation unit 3A, multiple positions of environments in cyberspace that are similar to the surrounding environment represented by the recognition data 8 may be detected for the same reasons as in estimating the real-space position of the vehicle P by the information processing device 2. In the example shown in FIG. 7, three candidate positions of the vehicle P in cyberspace, i.e., three candidate cyberspace positions of the vehicle P, are detected. Therefore, the estimation unit 3A associates each of the candidate cyberspace positions of the vehicle P with a cyber score calculated from each candidate cyberspace position of the vehicle P.

[0055] In this way, the estimation unit 3A estimates the cyberspace candidate position of the vehicle P at each time.

[0056] FIG. 8 is a diagram showing an example of estimation of a cyberspace candidate position of the vehicle P from the environment in cyberspace at each time shown in FIG.

[0057] In the example shown in FIG. 8, the cyberspace candidate position of vehicle P is estimated from time t-1 to time t+1, but the estimation period for the cyberspace candidate position of vehicle P is not limited to this. The start of the estimation period for the cyberspace candidate position of vehicle P may be time t-n1, and the end of the estimation period for the cyberspace candidate position of vehicle P may be time t+n2. n1 and n2 are each an integer greater than or equal to 0. n1 and n2 may be the same value or different values. The estimation period for the cyberspace candidate position of vehicle P is set by the user, and the set estimation period is modifiable.

[0058] The estimation unit 3A needs to estimate the cyberspace position of the vehicle P from the cyberspace environment at each time using at least one cyberspace candidate position of the vehicle P at each time. In this case, the estimation unit 3A estimates the cyberspace position of the vehicle P using not only the position information of the vehicle P but also the time direction information.

[0059] In the real world, the positions of vehicles and obstacles change over time, so the surrounding environment represented by the recognition data 8 also changes. Therefore, the cyberspace position of the vehicle P differs at each time. On the other hand, the amount of movement of the vehicle P at each time can be calculated from the measurement values ​​of various sensors 17 attached to the vehicle P.

[0060] Therefore, the estimated position obtained by adding the amount of movement of vehicle P during unit time Δt to the cyberspace candidate position of vehicle P at time t can be considered as the cyberspace candidate position of vehicle P at time t+1.

[0061] Figure 9 shows the cyberspace candidate position F of vehicle P at time t. t and the estimated position H of vehicle P at time t+1. t+1 FIG.

[0062] Considering these findings, among the cyberspace candidate positions of vehicle P at time t+1 after a unit time Δt has elapsed since a specific time t, a cyberspace candidate position of vehicle P that overlaps with an estimated position estimated from the amount of movement of vehicle P during unit time Δt is highly likely to be the cyberspace position of vehicle P at time t+1. Therefore, the estimation unit 3A classifies the cyberspace candidate positions of vehicle P at time t+1 that overlap with the estimated position into the same group as the cyberspace candidate positions of vehicle P at time t, which was the basis for calculating the estimated position. The estimation unit 3A performs the process of classifying the cyberspace candidate positions into groups in time series while sequentially changing the time t, thereby being able to generate groups of cyberspace candidate positions of vehicle P in time series.

[0063] Fig. 10 is a diagram showing an example in which cyberspace candidate positions of a vehicle P are classified into groups along a time series. In the example shown in Fig. 10, the cyberspace candidate positions of the vehicle P at each time are classified into three groups G1, G2, and G3 based on the cyberspace candidate positions of each vehicle P at time t-1. Fig. 10 is a schematic diagram that clearly shows the correspondence relationship between the cyberspace candidate positions of each vehicle P from time t-1 to time t+1 by shifting the cyberspace candidate positions of the vehicle P at time t and time t+1 horizontally so that the cyberspace candidate positions of the vehicles P included in the same group are aligned vertically as much as possible.

[0064] Note that the cyberspace candidate position of vehicle P overlaps with the estimated position means that even a portion of the cyberspace candidate position overlaps with the estimated position. Also, depending on the situation, there may be times when the estimated position does not overlap with the cyberspace candidate position of vehicle P, such as the estimated position at time t in group G3 in Figure 10. In such cases, the cyberspace candidate position of vehicle P at that time is treated as not existing. If the cyberspace candidate position of vehicle P does not exist, the cyber score associated with the cyberspace candidate position of vehicle P at that time will be "0".

[0065] For each generated group, the estimation unit 3A calculates the sum of the cyber scores associated with the cyberspace candidate positions of the vehicle P at each time within the group. The sum of the cyber scores represents the degree of agreement between the environment in cyberspace, taking into account not only the position direction of the vehicle P but also the time direction, and the surrounding environment represented by the recognition data 8.

[0066] Therefore, the estimation unit 3A estimates the cyberspace candidate position of the vehicle P at each time in the group with the largest total cyber score as the cyberspace position of the vehicle P at each time.

[0067] The correction unit 3B acquires the system time 30B and corrects the cyberspace position of the vehicle P estimated by the estimation unit 3A and the time at which the vehicle P is located at the position estimated by the estimation unit 3A based on the error in the position of the vehicle P estimated in real space and cyberspace, respectively, and the error between the edge time 30A and the system time 30B.

[0068] System time 30B is a reference time used in the position estimation system 1, and is, for example, the time set in the server 3. Preferably, the information processing device 2, server 3, and collection device 9 included in the position estimation system 1 each perform processing in synchronization with system time 30B. However, since there are multiple information processing devices 2 and collection devices 9 in the position estimation system 1, each device may perform processing according to the edge time 30A set in the device itself, for example, to reduce the amount of data communication in the communication network 5. Therefore, an error occurs between the edge time 30A and the system time 30B. The correction unit 3B calculates the time error between the edge time 30A and the system time 30B and determines the system time 30B corresponding to the time represented by the edge time 30A.

[0069] It is unlikely that an unnoticeable time difference will occur between the edge time 30A and the system time 30B over, for example, several hours. Therefore, the correction unit 3B does not need to calculate the time error every time edge data is received, but can calculate the time error every predetermined period. If the correction unit 3B associates the edge time 30A with the system time 30B using the most recent calculated time error, the amount of processing required for time correction can be reduced compared to when the time error is calculated every time edge data is received.

[0070] Furthermore, the estimation unit 3A estimates the cyberspace position of the vehicle P at each time based on the edge data. The environment in cyberspace and the surrounding environment represented by the recognition data 8 are constructed by aggregating measurement values ​​from the collection device 9 in the real space and various sensors 17 attached to the vehicle P, and therefore do not necessarily represent the actual environment.

[0071] If the edge data contains an error from the true value, the error in the edge data will affect the estimation of the cyberspace position of vehicle P, and the estimated cyberspace position of vehicle P will also contain an error. Furthermore, when estimating the cyberspace position of vehicle P in the next frame based on the cyberspace position of vehicle P estimated in the previous frame, the error in the cyberspace position of vehicle P in each frame up to that point will accumulate as the frame progresses, and the error in the estimated cyberspace position of vehicle P will become larger.

[0072] Therefore, the correction unit 3B corrects the cyberspace position of the vehicle P at each time estimated by the estimation unit 3A. To correct the cyberspace position of the vehicle P at each time, the cyberspace position of the vehicle P at each time, the cyber score associated with the cyberspace position of the vehicle P at each time, the real space position of the vehicle P at each time, and the edge score associated with the real space position of the vehicle P at each time are used.

[0073] Specifically, the correction unit 3B corrects the cyberspace position of the vehicle P at each time based on the ratio of the cyber score associated with the cyberspace position of the vehicle P to the edge score associated with the real-space position of the vehicle P at the same time.

[0074] The cyber score represents the reliability of the cyberspace position of the vehicle P estimated by the estimation unit 3A, and the edge score represents the reliability of the real space position of the vehicle P estimated by the edge estimation unit 2A of the information processing device 2. In other words, the position associated with a larger score represents the more accurate position of the vehicle P than the other position.

[0075] Therefore, correction unit 3B divides the distance between the cyberspace position of vehicle P and the real space position of vehicle P at each time t based on the ratio between the cyber score and the edge score, and corrects the cyberspace position of vehicle P so that the cyberspace position of vehicle P comes to the position of the division point. Note that correction unit 3B performs the division so that the division point approaches the larger of the cyber score and the edge score.

[0076] FIG. 11 is a diagram showing an example of correcting the cyberspace position of a vehicle P. For example, the cyberspace position L of the vehicle P at time t t The cyber score associated with the real space position R of the vehicle P at the same time t is 0.8. t The edge score associated with vehicle P is 0.2. t and the real space position R of vehicle P t The line segment D connecting these points is the cyberspace position L of the vehicle P. t and the real space position R of vehicle P t represents the distance from

[0077] In this case, the correction unit 3B calculates the division position based on the cyberspace position L of the vehicle P. t The cyberspace position of vehicle P at time t is corrected so that the cyberspace position of vehicle P is located at the division point that divides line segment D into 4:1 so that the cyberspace position of vehicle P is closer to the target.

[0078] In this way, the correction unit 3B corrects the cyberspace position of the vehicle P estimated by the estimation unit 3A using the edge score and the cyber score.

[0079] On the other hand, the communication unit 3C in FIG. 5 performs data communication with external devices connected to the communication network 5. For example, the communication unit 3C receives edge data from the information processing device 2 and the collection device 9, and transmits information related to the corrected cyberspace position of the vehicle P corrected by the correction unit 3B as cyber data to the information processing device 2. The cyber data includes, for example, the corrected cyberspace position of the vehicle P at each time, a cyber score associated with the corrected cyberspace position of the vehicle P at each time, and the cyberspace environment at each time constructed from the peripheral data. The communication unit 3C may transmit the error between the corrected cyberspace position of the vehicle P at each time and the real-space position of the vehicle P instead of or together with the corrected cyberspace position of the vehicle P.

[0080] The edge correction unit 2D of the information processing device 2 that receives the cyber data uses the cyberspace position and cyber score of the vehicle P received from the server 3 to correct the real-space position of the vehicle P at each time estimated by the edge estimation unit 2A.

[0081] The information processing device 2 and server 3 having these functions are configured, for example, by a computer. Fig. 12 is a diagram showing an example of the configuration of the information processing device 2 configured by a computer, and Fig. 13 is a diagram showing an example of the configuration of the server 3 configured by a computer.

[0082] 12, the information processing device 2 includes a control unit 10, a communication unit 12, a user interface unit 13 (hereinafter referred to as "UI unit 13"), an interface unit 14 (hereinafter referred to as "I / F unit 14"), and a storage unit 15. The control unit 10, the communication unit 12, the UI unit 13, the I / F unit 14, and the storage unit 15 are connected via a bus 16 so as to be able to exchange various types of information with one another.

[0083] The control unit 10 controls the operation of the information processing device 2. The control unit 10 is an example of a processor, and includes a CPU (Central Processing Unit) 10A, a ROM (Read Only Memory) 10B, and a RAM (Random Access Memory) 10C, which are responsible for processing of each functional unit of the information processing device 2 shown in Fig. 3. The ROM 10B stores in advance an information processing program 11 that the CPU 10A reads to estimate the real space position of the vehicle P, and various parameters that the CPU 10A references when controlling the operation of the information processing device 2. The RAM 10C is used as a temporary work area for the CPU 10A.

[0084] The communication unit 12 uses a predetermined communication protocol to perform wireless data communication with the server 3 connected to the communication network 5. Note that the communication destination of the communication unit 12 is not limited to the server 3, and data communication can be performed with any device connected to the communication network 5, that is, any external device.

[0085] The UI unit 13 receives instructions from the user and notifies the CPU 10A, and also notifies the user of various information processed by the information processing device 2. There are no restrictions on the means by which the UI unit 13 notifies the user of various information, and notification can be performed using at least one of the visual, auditory, and physical sensations.

[0086] The I / F unit 14 acquires measurement values ​​from various sensors 17 attached to the vehicle P wirelessly or via wired connections.

[0087] The storage unit 15 stores, for example, map data 4 of the area in which the vehicle P travels. The storage unit 15 is an example of a storage device that maintains stored information even if the power supplied to the storage unit 15 is cut off, and for example, a semiconductor memory such as an SSD (Solid State Drive) is used, but a hard disk may also be used.

[0088] 13, the server 3 includes a control unit 20, a communication unit 22, and a storage unit 23. The control unit 20, the communication unit 22, and the storage unit 23 are connected via a bus 24 so as to be able to exchange various information with each other.

[0089] The control unit 20 controls the operation of the server 3. The control unit 20 is an example of a processor, and includes a CPU 20A, a ROM 20B, and a RAM 20C that handle the processing of each functional unit of the server 3 shown in FIG. 5. The ROM 20B stores in advance a server program 21 that the CPU 20A reads to estimate the cyberspace position of the vehicle P, and various parameters that the CPU 20A references when controlling the operation of the server 3. The RAM 20C is used as a temporary work area for the CPU 20A.

[0090] The communication unit 22 uses a predetermined communication protocol to perform wireless data communication with the information processing device 2 and the collection device 9 connected to the communication network 5. Note that the communication destinations of the communication unit 22 are not limited to the server 3 and the collection device 9, and data communication can be performed with any external device connected to the communication network 5.

[0091] The storage unit 23 stores, for example, received edge data, etc. Similar to the storage unit 15 in the information processing device 2, the storage unit 23 is an example of a storage device in which stored information is maintained even if the power supplied to the storage unit 23 is cut off, and for example, a semiconductor memory such as an SSD is used, but a hard disk may also be used.

[0092] 12 and 13, respectively. Other units are connected to the bus 16 and the bus 24 of the server 3 as needed. For example, when a user issues an instruction to the server 3, a unit for notifying the CPU 20A of the instruction from the user, i.e., a unit having the same function as the UI unit 13 in the information processing device 2, is connected to the bus 24 of the server 3.

[0093] Next, the processing of the information processing device 2 will be described in detail. Fig. 14 is a flowchart showing an example of the flow of the estimation processing executed by the information processing device 2 when, for example, an instruction to start estimating the real space position of the vehicle P is received from the user. The CPU 10A of the information processing device 2 reads the information processing program 11 from the ROM 10B and executes the estimation processing. It is assumed that the map data 4 is stored in advance in the storage unit 15.

[0094] First, in step S10, the CPU 10A acquires the measurement values ​​of the various sensors 17 attached to the vehicle P through the I / F unit 14.

[0095] In step S20, CPU 10A generates recognition data 8 using the measurement values ​​of various sensors 17 acquired in the process of step S10.

[0096] In step S30, CPU 10A obtains map data 4 from storage unit 15.

[0097] In step S40, CPU 10A performs a matching process between recognition data 8 generated in the process of step S20 and map data 4 acquired in the process of step S30, and estimates the real space position of vehicle P.

[0098] In step S50, CPU 10A calculates an edge score at the real space position of vehicle P estimated by the process of step S40, and associates the calculated edge score with the real space position of vehicle P.

[0099] In step S60, the CPU 10A transmits to the server 3 as edge data the real-space position of the vehicle P, which is associated with the measurement values ​​of the various sensors 17 acquired by the processing of step S10, the recognition data 8 generated by the processing of step S20, and the edge score calculated by the processing of step S50.

[0100] In step S70, CPU 10A determines whether or not an instruction to end the estimation process has been received from the user, and if an instruction to end the estimation process has not been received, the process proceeds to step S80.

[0101] In step S80, CPU 10A determines whether or not a unit time Δt has elapsed since the measurement values ​​of the various sensors 17 were acquired by the process of step S10. If the unit time Δt has not elapsed, the determination process of step S80 is repeatedly executed and the process waits until the unit time Δt has elapsed. On the other hand, if the unit time Δt has elapsed, the process proceeds to step S10, and the measurement values ​​of the various sensors 17 for the next frame are acquired.

[0102] By repeatedly executing each process of steps S10 to S80 until the CPU 10A determines that an end instruction has been received from the user through the judgment process of step S70, the real space position of vehicle P at each time is estimated, and edge data for each time including the real space position of vehicle P is transmitted to server 3.

[0103] If it is determined in the determination process of step S70 that an end instruction has been received from the user, the estimation process shown in FIG. 14 ends.

[0104] It should be noted that not only the information processing device 2 but also the collection device 9 transmits peripheral data at each time to the server 3 as edge data.

[0105] Next, a detailed description will be given of the processing of the server 3. Fig. 15 is a flowchart showing an example of the flow of edge data accumulation processing executed by the server 3. The CPU 20A of the server 3 reads the server program 21 from the ROM 20B and executes the accumulation processing.

[0106] First, in step S100, the CPU 20A determines whether edge data has been received from the information processing device 2 and the collection device 9. If edge data has not been received, the CPU 20A repeatedly executes the determination process of step S100 to monitor the reception of edge data. If edge data has been received, the process proceeds to step S110.

[0107] In step S110, CPU 20A acquires the received edge data.

[0108] In step S120, the CPU 20A acquires the system time 30B and associates the system time 30B with the edge data acquired in the process of step S110.

[0109] In step S130, the CPU 20A stores the edge data associated with the system time 30B in the storage unit 23.

[0110] In step S140, CPU 20A determines whether an instruction to end the accumulation process has been received from the user. If an instruction to end the accumulation process has not been received, the process proceeds to step S100 and waits for the reception of edge data at the next time. By repeatedly executing the processes of steps S100 to S140 until CPU 20A determines in the determination process of step S140 that an instruction to end the accumulation process has been received from the user, edge data for each time is stored in storage unit 23.

[0111] On the other hand, if it is determined in the determination process of step S140 that an end instruction has been received from the user, the accumulation process shown in FIG. 15 is ended.

[0112] 16 is a flowchart showing an example of the flow of the estimation process executed by the server 3 when, for example, an instruction to start estimating the cyberspace position of the vehicle P is received from the user. The CPU 20A of the server 3 reads the server program 21 from the ROM 20B and executes the estimation process. It is assumed that the edge data for each time period has been stored in advance in the storage unit 23 by the accumulation process shown in FIG. 15. The accumulation process shown in FIG. 15 and the estimation process shown in FIG. 16 are executed in parallel, for example.

[0113] In step S200, the CPU 20A executes a candidate position estimation process for estimating a candidate position in cyberspace of the vehicle P at each time.

[0114] As already explained, there are cases where multiple cyberspace candidate positions of vehicle P are detected from the cyberspace environment at each time. Therefore, in step S300, CPU 20A classifies the cyberspace candidate positions of vehicle P into groups in chronological order, and calculates the total value of the cyber scores associated with each of the cyberspace candidate positions of vehicle P for each group, thereby executing a position estimation process to estimate the cyberspace position of vehicle P at each time.

[0115] In step S400, the CPU 20A executes a correction process to correct the cyberspace position of the vehicle P at each time estimated by the position estimation process, and the time when the vehicle P is present at the estimated position.

[0116] In step S500, the CPU 20A calculates the error between the corrected cyberspace position of the vehicle P and the real space position of the vehicle P at each time.

[0117] In step S600, CPU 20A transmits the cyber data including the corrected cyberspace position of vehicle P at each time corrected by the processing in step S400, the error between the corrected cyberspace position of vehicle P at each time and the real space position of vehicle P calculated in step S500, the cyber score associated with the corrected cyberspace position of vehicle P at each time, and the environment in cyberspace at each time to the information processing device 2 that transmitted the real space position of vehicle P. This completes the estimation process shown in FIG.

[0118] The CPU 10A of the information processing device 2 that receives the cyber data compares the cyber score associated with the cyberspace position of the vehicle P contained in the cyber data with the edge score associated with the real-space position of the vehicle P estimated by the CPU 10A, and corrects the real-space position of the vehicle P at each time.

[0119] If the edge score is equal to or greater than the cyber score, the real-space position of vehicle P estimated by the information processing device 2 represents the actual position of vehicle P with greater accuracy than, or with the same degree of accuracy as, the cyberspace position of vehicle P. Therefore, in this case, the CPU 10A of the information processing device 2 treats the estimated real-space position of vehicle P as the actual position of vehicle P, shares the received cyberspace environment, constructs an environment on RAM 10C that is the same as the cyberspace environment, and reflects the real-space position of vehicle P in the constructed environment.

[0120] On the other hand, if the edge score is smaller than the cyber score, the cyberspace position of vehicle P estimated by the server 3 represents the actual position of vehicle P with greater accuracy than the realspace position of vehicle P. Therefore, the CPU 10A of the information processing device 2 corrects the realspace position of vehicle P so that the estimated realspace position of vehicle P coincides with the cyberspace position of vehicle P, and reflects the realspace position of vehicle P in the environment constructed on RAM 10C by sharing the cyberspace environment.

[0121] That is, the information processing device 2 and the server 3 each refer to the other's position estimation result of the vehicle P and correct the position of the vehicle P estimated by their own device. Therefore, the accuracy of estimating the position of the vehicle P can be improved compared to when the information processing device 2 and the server 3 each estimate the position of the vehicle P independently.

[0122] Next, the candidate position estimation process executed in step S200 of FIG. 16 will be described.

[0123] 17 is a flowchart showing an example of the flow of candidate position estimation processing, which is executed for each edge data at each time.

[0124] In step S210, the CPU 20A constructs an environment in cyberspace using peripheral data included in the edge data at a specific time t stored in the storage unit 23.

[0125] In step S220, CPU 20A performs a matching process while moving the surrounding environment represented by the recognition data 8 included in the edge data at time t throughout the entire cyberspace environment constructed by the processing of step S210, and calculates the cyber score at each destination position.

[0126] In step S230, CPU 20A sets the position at which the largest cyber score is obtained from among the cyber scores calculated by the process of step S220 as the cyberspace candidate position of vehicle P.

[0127] As already explained, there may be multiple locations in cyberspace that are similar to the surrounding environment represented by the recognition data 8. Therefore, if the cyber scores at each location calculated by the processing of step S220 include a cyber score that is equal to or greater than a predetermined threshold, CPU 20A also sets the location where the cyber score calculated is equal to or greater than the predetermined threshold as a cyberspace candidate location of vehicle P.

[0128] The CPU 20A associates each of the cyberspace candidate positions of the vehicle P with a cyber score at the cyberspace candidate position.

[0129] The threshold value is set by, for example, a user and is stored in advance in the storage unit 23. The threshold value is set to a value that, if equal to or greater than this value, is preferable as a cyberspace candidate position of the vehicle P.

[0130] This completes the candidate position estimation process shown in Fig. 17. By executing the candidate position estimation process shown in Fig. 17 using the edge data at each time, the cyberspace candidate position of the vehicle P at each time is estimated.

[0131] FIG. 18 is a flowchart showing an example of the flow of the position estimation process executed by the process of step S300 in FIG. 16 following the candidate position estimation process.

[0132] In step S310, the CPU 20A uses the measurement values ​​of the various sensors 17 included in the edge data to obtain the amount of movement of the vehicle P from a specific time t until a unit time Δt has elapsed.

[0133] In step S320, CPU 20A adds the movement amount of vehicle P obtained by the processing of step S310 to the cyberspace candidate position of each vehicle P at time t, and calculates the estimated position of vehicle P in the next frame relative to the cyberspace candidate position of each vehicle P.

[0134] In step S330, CPU 20A classifies the cyberspace candidate positions of vehicle P in the next frame that overlap with the estimated position calculated by the processing of step S320 into the same group as the cyberspace candidate positions of vehicle P at time t from which the estimated position was calculated.

[0135] In step S340, CPU 20A determines whether grouping of cyberspace candidate positions of vehicle P has been performed for all frames. If there is a frame for which grouping has not been performed, the process proceeds to step S310, where the amount of movement of vehicle P from the next frame until unit time Δt has elapsed is obtained. That is, CPU 20A generates groups of cyberspace candidate positions of vehicle P in chronological order by repeatedly performing the processes of steps S310 to S340 until it is determined by the determination process of step S340 that grouping of cyberspace candidate positions of vehicle P for all frames has been completed.

[0136] If it is determined in the determination process of step S340 that grouping of the cyberspace candidate positions of the vehicle P for all frames has been completed, the process proceeds to step S350.

[0137] In step S350, the CPU 20A calculates, for each group generated by the processes of steps S310 to S340, the total value of the cyber scores associated with the cyberspace candidate positions of the vehicle P at each time within the group.

[0138] In step S360, CPU 20A estimates the cyberspace candidate position of vehicle P at each time in the group with the largest total cyber score as the cyberspace position of vehicle P at each time.

[0139] This completes the position estimation process shown in Fig. 18. By the position estimation process shown in Fig. 18, the cyberspace position of the vehicle P is estimated using information on the position direction and time direction of the vehicle P.

[0140] FIG. 19 is a flowchart showing an example of the flow of the correction process executed in step S400 of FIG. 16 following the position estimation process.

[0141] In step S410, the CPU 20A calculates the time error between the edge time 30A included in the edge data used to estimate the cyberspace position of the vehicle P and the system time 30B associated with the edge data by the processing of step S120 in Figure 15.

[0142] In step S420, the CPU 20A acquires, from the edge data stored in the storage unit 23, the real space position of the vehicle P at each time and the edge score associated with the real space position.

[0143] In step S430, CPU 20A acquires the cyberspace position of vehicle P at each time estimated by the position estimation process in step S300 shown in FIG. 16, and the cyber score associated with the cyberspace position.

[0144] In step S440, CPU 20A divides the distance between the cyberspace position of vehicle P and the real space position of vehicle P at the same time based on the ratio of the corresponding cyber score and edge score, and corrects the cyberspace position of vehicle P at each time so that the cyberspace position of vehicle P is at the position of the division point.

[0145] Note that if the edge score used to correct the cyberspace position of vehicle P is too low, correcting the cyberspace position of vehicle P using the edge score may not contribute to improving the estimation accuracy of the cyberspace position of vehicle P. Therefore, if the edge score used to correct the cyberspace position of vehicle P is below the lower limit value that is considered to contribute to improving the estimation accuracy of the cyberspace position of vehicle P, it is preferable that CPU20A use the cyberspace position of vehicle P at each time as the corrected cyberspace position of vehicle P as is.

[0146] The corrected cyberspace position of vehicle P is not the position at edge time 30A indicated by the edge data including the edge score used to correct the cyberspace position of vehicle P, but the cyberspace position of vehicle P at the time taking into account the time error calculated by the processing of step S410.

[0147] The process of step S410 need not be performed every time the correction process is executed, but may be performed at a predetermined period, for example, once a month.

[0148] As a result, the correction process shown in FIG. 19 ends, and CPU 20A executes step S500 and subsequent steps of the estimation process shown in FIG.

[0149] In this way, the position estimation system 1 of the present disclosure creates an environment in cyberspace in which the vehicle P travels using surrounding data collected by the collection device 9, and estimates the position of the vehicle P in cyberspace based on the recognition data 8 of the vehicle P. Therefore, in a situation in which the vehicle P travels in a place where visibility is poor due to obstacles, the accuracy of estimating the position of the vehicle P can be improved compared to when the vehicle P estimates its own position by itself.

[0150] Furthermore, even if there are multiple candidate cyberspace positions of the vehicle P at a specific time, the position estimation system 1 of the present disclosure can narrow down the cyberspace position of the vehicle P to one based on the positional relationship with the candidate cyberspace positions of the vehicle P at other times. Therefore, for example, in a situation where the vehicle P is traveling on a straight road with no landmarks such as buildings in the vicinity, the accuracy of estimating the position of the vehicle P can be improved compared to when the vehicle P estimates its own position by itself.

[0151] Furthermore, the position estimation system 1 of the present disclosure corrects the estimated cyberspace position of the vehicle P using the real-space position and edge score of the vehicle P. Therefore, the accuracy of estimating the position of the vehicle P can be improved compared to when the cyberspace position of the vehicle P is estimated using only information obtained from the cyberspace environment.

[0152] While one embodiment of the position estimation system 1 for the vehicle P has been described above, the disclosed embodiment of the position estimation system 1 is merely an example, and the embodiment of the position estimation system 1 is not limited to the scope described in the embodiment. Various modifications or improvements can be made to the embodiment without departing from the gist of the present disclosure, and the modified or improved form of the position estimation system 1 is also included in the technical scope of the disclosure.

[0153] For example, the internal processing order in the flowcharts shown in FIGS. 14 to 19 may be changed without departing from the scope of the present disclosure.

[0154] In the above embodiment, the processes shown in Figures 14 to 19 are implemented by software. However, the processes equivalent to those shown in the flowcharts of the processes may be executed by hardware. In this case, the processes can be executed at a higher speed than when the processes are implemented by software.

[0155] In the above embodiment, an example has been described in which the information processing program 11 is stored in ROM 10B and the server program 21 is stored in ROM 20B. However, the storage destinations of the information processing program 11 and the server program 21 are not limited to ROM 10B and ROM 20B, respectively. The information processing program 11 and the server program 21 can also be provided in a form recorded on a computer-readable storage medium.

[0156] For example, the information processing program 11 and the server program 21 may be provided in a form recorded on an optical disk such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a Blu-ray disc. The information processing program 11 and the server program 21 may also be provided in a form recorded on a portable semiconductor memory such as a USB (Universal Serial Bus) memory or a memory card. ROM 10B, ROM 20B, CD-ROM, DVD-ROM, Blu-ray disc, USB, and memory card are examples of non-transitory storage media.

[0157] Furthermore, the CPU 10A of the information processing device 2 may download the information processing program 11 from an external device connected to the communication network 5 through the communication unit 12. Similarly, the CPU 20A of the server 3 may download the server program 21 from an external device connected to the communication network 5 through the communication unit 22.

[0158] The controller and methods described herein may be implemented by a special-purpose computer having a processor programmed to perform one or more functions embodied in a computer program. Alternatively, the apparatus and methods described herein may be implemented by a special-purpose computer having a processor configured with dedicated hardware logic circuitry. Alternatively, the apparatus and methods described herein may be implemented by one or more special-purpose computers configured by a combination of a processor executing a computer program and one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by a computer on a computer-readable non-transitory storage medium.

[0159] The following are notes related to this disclosure.

[0160] (Appendix 1) an estimation unit (3A) that estimates a position of a vehicle (P) at each time in cyberspace using edge data at each time including information on the position of the vehicle (P) estimated in real space and the environment in which the vehicle is traveling; a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on errors in the position and time of the vehicle in real space and cyberspace; a communication unit (3C) that transmits information about the vehicle position corrected by the correction unit to an information processing device (2) that estimated the vehicle position in real space; A server (3) equipped with:

[0161] (Appendix 2) the estimation unit estimates at least one candidate position of the vehicle in cyberspace for each time using a cyber score that indicates a degree of agreement between an environment in cyberspace at each time in which the vehicle is traveling, which is constructed based on the edge data, and a surrounding environment at the position of the vehicle estimated in real space; grouping the estimated candidate positions of the vehicle in time series, and calculating a total value of the cyber scores associated with the candidate positions of the vehicle at each time for each group; The candidate position of the vehicle at each time in the group with the largest total cyber score is estimated as the position of the vehicle in cyberspace at each time. The server described in Appendix 1.

[0162] (Appendix 3) the estimation unit calculates an estimated position by adding, to each of the candidate positions of the vehicle at a specific time, an amount of movement of the vehicle during an estimation interval (Δt) between the positions of the vehicle in real space; Among the candidate positions of the vehicle at a time after the elapse of the estimation interval from the specific time, the candidate positions of the vehicle that overlap with the estimated position are classified into the same group as the candidate positions of the vehicle at the specific time from which the estimated position was calculated, while sequentially changing the specific time along a time series, thereby generating groups of candidate positions of the vehicle along a time series. The server described in Appendix 2.

[0163] (Appendix 4) The correction unit corrects the position of the vehicle at each time in cyberspace estimated by the estimation unit using the position of the vehicle at each time estimated in real space and an edge score, which is information included in the edge data and indicates the degree of agreement between the environment of the space in which the vehicle is traveling, obtained from pre-prepared map data (4), and the surrounding environment at the position of the vehicle at each time estimated in real space. 2. A server according to claim 2 or claim 3.

[0164] (Appendix 5) The correction unit corrects the position of the vehicle estimated in cyberspace at each time based on a ratio between the cyber score associated with the position of the vehicle in cyberspace at each time and the edge score associated with the position of the vehicle in real space at each time. A server as described in Appendix 4.

[0165] (Appendix 6) an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that indicates a degree of coincidence between an environment on the map data at the position of the vehicle estimated by the edge estimation unit and a surrounding environment recognized by the vehicle, and associates the calculated edge score with the position of the vehicle for each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits the vehicle position estimated at each time and information about the surrounding environment at each vehicle position, including the edge score and the measurement value of the sensor, as edge data to a server (3) that configures an environment identical to real space in cyberspace; an edge correction unit (2D) that corrects the position of the vehicle estimated by the edge estimation unit using information regarding the position of the vehicle in cyberspace received from the server by the edge communication unit; An information processing device (2) equipped with the above.

[0166] (Appendix 7) On the computer, Using edge data for each time including the position of a vehicle (P) estimated in real space and information about the environment in which the vehicle is traveling, the position of the vehicle is estimated for each time in cyberspace; correcting the estimated position of the vehicle in cyberspace and the time at which the vehicle is present at the estimated position based on errors in the position and time of the vehicle in real space and cyberspace; and executing a process of transmitting information about the corrected position of the vehicle to an information processing device (2) that estimated the position of the vehicle in real space. Server program (21).

[0167] (Appendix 8) On the computer, The position of the vehicle (P) in the real space at each time is estimated using the measurement values ​​of the sensor (17) obtained at each time and the map data (4), calculating an edge score representing a degree of coincidence between an environment on the map data at the estimated position of the vehicle and a surrounding environment recognized by the vehicle, and associating the calculated edge score with the position of the vehicle at each estimated time; The position of the vehicle estimated at each time point and information on the surrounding environment at each position of the vehicle, including the edge score and the measurement value of the sensor, are transmitted as edge data to a server (3) configured in cyberspace to create an environment identical to real space; and executing a process of correcting the estimated position of the vehicle using information on the position of the vehicle in cyberspace received from the server. Information Processing Programs (11).

[0168] (Appendix 9) an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that indicates a degree of coincidence between an environment on the map data at the position of the vehicle estimated by the edge estimation unit and a surrounding environment recognized by the vehicle, and associates the calculated edge score with the position of the vehicle for each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits the vehicle position estimated at each time and information about the surrounding environment at each vehicle position, including the edge score and the measurement value of the sensor, as edge data to a server (3) that configures an environment identical to real space in cyberspace; an edge correction unit (2D) that corrects the position of the vehicle estimated by the edge estimation unit using information regarding the position of the vehicle in cyberspace received from the server by the edge communication unit; An information processing device (2) comprising: an estimation unit (3A) that estimates a position of the vehicle at each time in cyberspace using the edge data at each time including information on the position of the vehicle estimated in real space and information on the environment in which the vehicle travels, received from the information processing device; a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on errors in the position and time of the vehicle in real space and cyberspace; a communication unit (3C) that transmits information about the vehicle position corrected by the correction unit to the information processing device (2); Server with (3) A location estimation system (1) including:

[0169] (Appendix 10) A non-transitory storage medium (20B) storing a server program (21) executable by a computer for executing an estimation process, The estimation process a step of estimating a position of a vehicle (P) at each time in a cyberspace using edge data at each time including information on the position of the vehicle (P) estimated in real space and the environment in which the vehicle is traveling; correcting the estimated position of the vehicle in cyberspace and the time at which the vehicle is present at the estimated position based on errors in the position and time of the vehicle in real space and cyberspace; transmitting information about the corrected position of the vehicle to an information processing device (2) that estimated the position of the vehicle in real space; Non-transitory storage media, including:

[0170] (Appendix 11) A non-transitory storage medium (10B) storing an information processing program (11) executable by a computer for executing an estimation process, The estimation process a step of estimating the position of the vehicle (P) at each time in the real space using measurement values ​​of the sensor (17) obtained at each time and map data (4); calculating an edge score representing a degree of coincidence between an environment on the map data at the estimated position of the vehicle and a surrounding environment recognized by the vehicle, and associating the calculated edge score with the position of the vehicle at each estimated time; transmitting, as edge data, the vehicle positions estimated at each time point and information on the surrounding environment at each vehicle position, including the edge scores and the measurement values ​​of the sensors, to a server (3) configured in cyberspace to represent an environment identical to real space; correcting the estimated position of the vehicle using information about the position of the vehicle in cyberspace received from the server; Non-transitory storage media, including: [Explanation of symbols]

[0171] 1 Position estimation system, 2 Information processing device, 2A Edge estimation unit, 2B Edge calculation unit, 2C Edge communication unit, 2D Edge correction unit, 2E Memory unit, 3 Server, 3A Estimation unit, 3B Correction unit, 3C Communication unit, 4 Map data, 5 Communication network, 6A Road in real space, 6B Road in cyberspace, 7 Wireless communication device, 8 Recognition data, 9 Collection device, 10 Control unit of information processing device, 10A CPU of information processing device, 11 Information processing program, 17 Various sensors, 20 Server control unit, 20A Server CPU, 21 Server program, 30A Edge time, 30B System time, Δt Unit time, G1, G2, G3 Group, P, Q Vehicle, t Time

Claims

1. an estimation unit (3A) that estimates a position of a vehicle (P) at each time in cyberspace using edge data at each time including information on the position of the vehicle (P) estimated in real space and the environment in which the vehicle is traveling; a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on errors in the position and time of the vehicle in real space and cyberspace; a communication unit (3C) that transmits information about the vehicle position corrected by the correction unit to an information processing device (2) that estimated the vehicle position in real space; A server (3) comprising:

2. the estimation unit estimates at least one candidate position of the vehicle in cyberspace for each time using a cyber score that indicates a degree of agreement between an environment in cyberspace at each time in which the vehicle is traveling, which is constructed based on the edge data, and a surrounding environment at the position of the vehicle estimated in real space; grouping the estimated candidate positions of the vehicle in time series, and calculating a total value of the cyber scores associated with the candidate positions of the vehicle at each time for each group; The candidate position of the vehicle at each time in the group with the largest total cyber score is estimated as the position of the vehicle in cyberspace at each time. The server of claim 1 .

3. the estimation unit calculates an estimated position by adding, to each of the candidate positions of the vehicle at a specific time, an amount of movement of the vehicle during an estimation interval (Δt) between the positions of the vehicle in real space; Among the candidate positions of the vehicle at a time after the elapse of the estimation interval from the specific time, the candidate positions of the vehicle that overlap with the estimated position are classified into the same group as the candidate positions of the vehicle at the specific time from which the estimated position was calculated, while sequentially changing the specific time along a time series, thereby generating groups of candidate positions of the vehicle along a time series. The server of claim 2.

4. The correction unit corrects the position of the vehicle at each time in cyberspace estimated by the estimation unit using the position of the vehicle at each time estimated in real space and an edge score, which is information included in the edge data and represents a degree of agreement between the environment of the space in which the vehicle travels, obtained from pre-prepared map data (4), and the surrounding environment at the position of the vehicle at each time estimated in real space. The server according to claim 2 or claim 3.

5. The correction unit corrects the position of the vehicle estimated in cyberspace at each time based on a ratio between the cyber score associated with the position of the vehicle in cyberspace at each time and the edge score associated with the position of the vehicle in real space at each time. The server of claim 4.

6. an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that indicates a degree of coincidence between an environment on the map data at the position of the vehicle estimated by the edge estimation unit and a surrounding environment recognized by the vehicle, and associates the calculated edge score with the position of the vehicle for each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits, as edge data, the vehicle position estimated at each time and information on the surrounding environment at each vehicle position including the edge score and the measurement value of the sensor, to a server (3) that configures an environment identical to real space in cyberspace; an edge correction unit (2D) that corrects the position of the vehicle estimated by the edge estimation unit using information regarding the position of the vehicle in cyberspace received from the server by the edge communication unit; An information processing device (2) comprising:

7. On the computer, Using edge data for each time including the position of a vehicle (P) estimated in real space and information about the environment in which the vehicle is traveling, the position of the vehicle is estimated for each time in cyberspace; correcting the estimated position of the vehicle in cyberspace and the time at which the vehicle is present at the estimated position based on errors in the position and time of the vehicle in real space and cyberspace; and transmitting information about the corrected position of the vehicle to an information processing device (2) that estimated the position of the vehicle in real space. Server program (21).

8. On the computer, The position of the vehicle (P) in the real space at each time is estimated using the measurement values ​​of the sensor (17) obtained at each time and the map data (4), calculating an edge score representing a degree of coincidence between an environment on the map data at the estimated position of the vehicle and a surrounding environment recognized by the vehicle, and associating the calculated edge score with the position of the vehicle at each estimated time; The vehicle positions estimated at each time point and information on the surrounding environment at each vehicle position, including the edge scores and the measurement values ​​of the sensors, are transmitted as edge data to a server (3) configured in cyberspace to create an environment identical to that of real space; and executing a process of correcting the estimated position of the vehicle using information on the position of the vehicle in cyberspace received from the server. Information processing program (11).

9. an edge estimation unit (2A) that estimates the position of a vehicle (P) at each time in real space using measurement values ​​of a sensor (17) obtained at each time and map data (4); an edge calculation unit (2B) that calculates an edge score that indicates a degree of coincidence between an environment on the map data at the position of the vehicle estimated by the edge estimation unit and a surrounding environment recognized by the vehicle, and associates the calculated edge score with the position of the vehicle for each time estimated by the edge estimation unit; an edge communication unit (2C) that transmits, as edge data, the vehicle position estimated at each time and information on the surrounding environment at each vehicle position including the edge score and the measurement value of the sensor, to a server (3) that configures an environment identical to real space in cyberspace; an edge correction unit (2D) that corrects the position of the vehicle estimated by the edge estimation unit using information regarding the position of the vehicle in cyberspace received from the server by the edge communication unit; An information processing device (2) comprising: an estimation unit (3A) that estimates a position of the vehicle at each time in cyberspace using the edge data at each time including information on the position of the vehicle estimated in real space and information on the environment in which the vehicle travels, received from the information processing device; a correction unit (3B) that corrects the position of the vehicle in cyberspace estimated by the estimation unit and the time at which the vehicle is present at the position estimated by the estimation unit based on errors in the position and time of the vehicle in real space and cyberspace; a communication unit (3C) that transmits information about the vehicle position corrected by the correction unit to the information processing device (2); A server (3) equipped with A location estimation system (1) comprising:

Images