Vehicle position calculation method and vehicle position calculation system

Abstract

To provide a vehicle position calculation technique capable of calculating a vehicle position with accuracy equal to or higher than the resolution of a sensor.SOLUTION: An identifier 205 having a first plane and a second plane substantially perpendicular to a rail surface and crossing each other is set. Then, the first plane and the second plane of the identifier are irradiated with light from the sensing unit 108 of the vehicle, and a first point cloud 206A and a second point cloud 206B are extracted as a set of reflected light from each of the first plane and the second plane of the identifier. A relative position of the identifier with respect to the vehicle is calculated based on an intersection of a first straight line calculated from the first point group and a second straight line calculated from the second point group.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a vehicle position calculation method and a vehicle position calculation system. [Background technology]

[0002] In recent years, automatic train operation (ATO) devices have been increasingly introduced into train operations in order to reduce the burden on train crews and cut labor costs. The ATO device is a device that transmits location information from a ground terminal installed on the ground within the tracks to a device called an on-board terminal installed on the train, and estimates the current location of the train based on the location information received by the on-board terminal. While location estimation methods using ground coils have the advantages of being highly environmentally resistant and having high accuracy in estimating the current location, they have issues such as high equipment costs for the ground coils and on-board coils, as well as high installation and maintenance costs for the ground coils. On the other hand, with train schedules becoming increasingly congested and platform doors being installed, there is a need for stricter stopping locations.

[0003] Patent Document 1 discloses the following technology as a train position detection device, with the objective of detecting the position of a train inexpensively and reliably. "The optical radar of the train equipment has a light emitting means capable of emitting light at different irradiation angles relative to the direction of train travel onto a predetermined area on a plane ahead of the train, and a light receiving means capable of receiving light reflected from an object when the irradiated light is reflected. The outer shape of an object ahead of the train is determined based on the light irradiation angle and the time between emitting the light and receiving the reflected light (reflection time). The memory means has an object information database in which the outer shapes and reference points of objects along the track are stored, and a position information database in which position reference information including object identification information, the direction of the object's reference point, and the distance to the reference point at each position along the track is stored in correspondence with the position information. The position of the train is detected based on the outer shape of the determined object, the object information database, and the position information database." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-024521 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the invention of Patent Document 1, the relative distance calculation resolution depends on the performance of the sensor, and it is difficult to improve the vehicle position calculation accuracy beyond the accuracy determined by the sensor resolution. Therefore, an object of the present invention is to provide a vehicle position calculation technique that can calculate the vehicle position with an accuracy equal to or greater than the resolution of the sensor. [Means for solving the problem]

[0006] To solve the above problem, one representative vehicle position calculation technique of the present invention sets an identifier having a first plane and a second plane that are approximately perpendicular to the rail surface and intersect with each other. Then, a first point cloud and a second point cloud are extracted as collections of reflected light from the first plane and the second plane of the identifier, respectively. Then, the relative position of the identifier with respect to the vehicle is calculated based on the intersection of a first straight line calculated from the first point cloud and a second straight line calculated from the second point cloud. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a vehicle position calculation technique that can calculate the vehicle position with an accuracy equal to or greater than the resolution of the sensor. Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiments of the invention. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing a train stopping at a station. [Figure 2] FIG. 2 is a diagram illustrating an example of the overall configuration of an automatic train operation system. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of the vehicle position calculation system according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of an identifier installed near a railroad track. [Figure 5] FIG. 5 is a diagram illustrating an example of information included in an identifier. [Figure 6] FIG. 6 is a diagram showing an example of the first point cloud, the second point cloud, and the point cloud of the rail. [Figure 7] FIG. 7 is a diagram showing how the sensing unit irradiates the identifier with laser light. [Figure 8] FIG. 8 is a diagram showing derivation of a rail surface from a point cloud of a rail. [Figure 9] FIG. 9 is a diagram showing a point cloud projected onto the rail surface at a second time. [Figure 10] FIG. 10 is a diagram showing a state in which the first point cloud acquired at a first time and the second point cloud are superimposed. [Figure 11] FIG. 11 is a diagram showing the intersection of the first line and the second line. [Figure 12] FIG. 12 is a flowchart showing the position estimation process executed by the sensor data processing unit. [Figure 13] FIG. 13 is a diagram showing acquisition of a point cloud of identifiers at a distant location. [Figure 14] FIG. 14 is a diagram showing acquisition of a point cloud of identifiers at an intermediate point. [Figure 15] FIG. 15 is a diagram showing acquisition of a point cloud of identifiers at nearby points. [Figure 16] FIG. 16 is a diagram showing a point cloud projected onto the rail surface of the identifier at the first location. [Figure 17] FIG. 17 is a diagram showing a point cloud projected onto the rail surface of the identifier at the second location. [Figure 18] FIG. 18 is a diagram showing a point cloud that is projected onto the rail surface of the identifier and stored. [Figure 19] FIG. 19 is a diagram showing a case where the first and second planes that constitute the rail are not perpendicular to the rail surface. [Figure 20] FIG. 20 is a diagram showing a point cloud of the identifiers projected onto the rail surface in FIG. [Figure 21] Figure 21 is a diagram in which railway signs and platform doors are used as identifiers. [Figure 22] FIG. 22 is a diagram illustrating the intersection of the first line and the second line in the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, examples of the present disclosure will be described with reference to the drawings. Note that the present disclosure is not limited to these examples. In addition, in the description of the drawings, the same parts are designated by the same reference numerals. When there are multiple components with the same or similar functions, they may be described using the same reference numeral with different subscripts. When there is no need to distinguish between these multiple components, the subscripts may be omitted. Furthermore, although terms such as "first," "second," and "third" may be used to describe various elements or components in the present disclosure, it will be understood that these elements or components should not be limited by these terms. These terms are used only to distinguish one element or component from another. Thus, a first element or component discussed below could also be referred to as a second element or component without departing from the teachings of the concepts of the present disclosure. To facilitate understanding of the disclosure, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present disclosure is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

[0010] <Background> First, the background art will be described with reference to FIG. FIG. 1 is a diagram showing a vehicle 107 stopping at a station. As shown in FIG. 1, the vehicle 107 may stop at a station equipped with platform doors. At such stations, the vehicle 107 must accurately align the position of the boarding gate with the platform doors, and high accuracy in the stopping position is required. To do this, it is first necessary to accurately determine the position of the vehicle.

[0011] <Automatic train operation system> Next, the overall configuration of the automatic train operation system 100, which is the premise of this disclosure, will be described with reference to FIG. FIG. 2 is a diagram showing the overall configuration of the automatic train operation system 100. The automatic train operation system 100 mounted on a vehicle 107 includes a speed generator 101, a speed calculation device 102, a position estimation device 103, an automatic train operation device 104, a vehicle information control device 105, and a braking / driving control device 106. Although the following description is based on the premise of a train, the technology of the present disclosure is not limited to trains and may be applied to other moving bodies such as automobiles, buses, monorails, etc.

[0012] These devices may be configured in any form as long as they can perform the functions described below. For example, the automatic train operation system 100 may be implemented as independent hardware, or as a software module executed by a CPU, memory, and RAM (not shown) in the automatic train operation system 100.

[0013] <Telecogenerator> The speed generator 101 is installed on the wheel axle and can detect the rotation speed of the wheels of the automatic train operation system 100. The detected rotation speed is transmitted to the speed calculation device 102 as a speed signal.

[0014] <Speed ​​calculation device> The speed calculation device 102 acquires a speed signal from the speed generator 101, and is able to calculate the vehicle speed from the speed signal and the wheel diameter.

[0015] <Position estimation device> The position estimation device 103 is a device that estimates its own position information from the position information of one or more identifiers installed around the line and information on the relative positions (relative distances) between each identifier and the train. If the identifier cannot be recognized, the position estimation device 103 may calculate its own position from the position information of the departure station and the integrated value of the vehicle speed.

[0016] <Automatic train operation device> The automatic train operation device 104 includes a planning unit (not illustrated) having a planning function and a following unit (not illustrated) having a following function. Here, the planning function is a function of calculating a target speed by checking the current vehicle position against a previously stored vehicle speed pattern up to the station stop position. The tracking function is a function that inputs the speed deviation between the target speed and the current vehicle speed and calculates the braking / driving force to be output. The automatic train operation device 104 can include the calculated braking / driving force in a control command and output it to the vehicle information control device 105.

[0017] <Vehicle information control device> The vehicle information control device 105 can control the automatic train operation system 100 and can transmit control commands output by the automatic train operation device 104 to the braking / driving control device 106.

[0018] <Braking / driving control device> The braking / driving control device 106 can drive the wheel axles based on the control command sent from the vehicle information control device 105, and can cause the vehicle 107 to run or stop. [Example]

[0019] Next, the configuration of the vehicle position calculation system 200 according to the first embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of the configuration of a vehicle position calculation system 200 according to the first embodiment. The vehicle position calculation system 200 can calculate the own position of the vehicle 107 and output the own position to a control command calculation unit of an automatic train operation device (not shown). The system also includes a sensing unit 108, a sensor data processing unit 202, an identifier point cloud accumulation database 112, an identifier information database 114, and a vehicle position calculation unit 115, and is controlled by a control unit 116.

[0020] Furthermore, the vehicle position calculation system 200 is installed on the vehicle 107 and is supplied with power from an electrical wiring facility provided along the rail 120 . Furthermore, the relative position between the vehicle 107 and the identifier 205 can be calculated from the identifiers 205 installed around the route and the relative positions between the identifiers 205 and the vehicle 107 . The identifier 205 indicates the side of the route (left side, right side, or both), on the route, within the route, etc., and the location or position at which it is installed is not particularly limited as long as it is within a range that can be easily calculated by the sensing unit 108.

[0021] <Sensing section> The sensing unit 108 can, for example, use LiDAR (Light Detection and Ranging) to scan the identifier 205 and the rail, measure the two-dimensional or three-dimensional shape of the identifier 205 and the rail, and transmit the shape data (measurement data) to the identifier recognition unit 109 and the rail recognition unit 110.

[0022] Here, LiDAR is a method of measuring the distance to an object by irradiating the object with laser light and receiving the reflected light. The distance measurement accuracy of LiDAR is relatively high, ranging from a few mm to a few cm, and by using a single straight-line laser as the basic unit and periodically changing the direction of irradiation at a high speed of several tens of Hz, the surface shape of the irradiated object can be captured as a collection of minute points (point cloud). Therefore, the aforementioned measurement data is data (data group) that expresses the two-dimensional or three-dimensional shape of the identifier 205 as a point cloud. For example, in the embodiment of the present disclosure, the sensing unit 108 can irradiate a first plane and a second plane, which will be described later, with laser light and acquire reflected light.

[0023] In the embodiment of the present disclosure, the LiDAR of the sensing unit 108 may be a three-dimensional LiDAR that can change the direction of irradiation three-dimensionally, or may be a two-dimensional LiDAR that can change the direction of irradiation two-dimensionally. Furthermore, a three-dimensional LiDAR and a two-dimensional LiDAR may be used alone or in combination. Similarly, if the surface shape of the object can be captured, the data acquired by the sensing unit 108 may be data obtained by a stereo camera, for example.

[0024] <Sensor data processing unit> The sensor data processing unit 202 includes an identifier recognition unit 109, a rail recognition unit 110, an identifier point cloud position calculation unit 111, and an identifier point cloud matching unit 113, and acquires position information (latitude, longitude, altitude) of the identifier 205 from the measurement data acquired by the sensing unit 108, and calculates the relative distance between each identifier 205 and the vehicle 107.

[0025] Thereafter, the sensor data processing unit 202 transmits the calculated relative distance between the vehicle 107 and the identifier 205 and the position information of the identifier 205 to the vehicle position calculation unit 115 .

[0026] <<Identifier recognition unit, rail recognition unit>> The identifier recognition unit 109 can extract a first point cloud 206A, which is a collection of points of reflected light from a first plane, and a second point cloud 206B, which is a collection of points of reflected light from a second plane, from the measurement data acquired from the sensing unit 108. Similarly, the rail recognition unit 110 can extract, from the measurement data acquired from the sensing unit 108, a rail point cloud 206C that is a collection of points of light reflected from the top surface of the rail. Furthermore, the rail recognition unit 110 can calculate the rail surface based on the extracted rail point cloud 206C. Here, the rail surface is the surface formed by the upper left and right surfaces of the rail 120 on which the vehicle 107 runs. Furthermore, the first point cloud 206A, the second point cloud 206B, the rail point cloud 206C, and the rail surface are transmitted to the identifier point cloud position calculation unit 111.

[0027] The sensing unit 108, the identifier recognition unit 109, and the rail recognition unit 110 may be integrated into a general-purpose sensor. In this way, by using the same general-purpose sensor to perform both the function of acquiring the information contained in the identifier 205 and the function of calculating the relative distance between the identifier 205 and the vehicle 107, the cost of the sensor can be reduced, and the process of estimating the position of the vehicle 107 can be achieved at low cost.

[0028] On the other hand, the sensing unit 108, the identifier recognition unit 109, and the rail recognition unit 110 may be separate units. In this way, by using separate dedicated sensors to perform the functions of acquiring information contained in identifier 205 and calculating the relative distance / relative attitude between identifier 205 and vehicle 107, it is possible to use sensors specialized for each function, thereby improving the accuracy of estimating the position of vehicle 107.

[0029] <<Identifier point cloud position calculation unit>> The identifier point cloud position calculation unit 111 projects the first point cloud 206A and the second point cloud 206B onto the rail surface and transmits the positions of the first point cloud 206A and the second point cloud 206B on the rail surface to the identifier point cloud matching unit 113.

[0030] <<Identifier point cloud matching part>> The identifier point cloud matching unit 113 matches the positions of the first point cloud 207A and the second point cloud 207B acquired at a first time and stored in the identifier point cloud storage database 112 with the positions of the first point cloud 206A and the second point cloud 206B acquired at a second time and projected onto the rail surface. If, as a result of matching, there are similarities or matching points between the first point cloud 207A and the second point cloud 207B stored in the identifier point cloud accumulation database 112 and the first point cloud 206A and the second point cloud 206B, the data of the first point cloud 206A and the second point cloud 206B acquired at the second time are superimposed and accumulated in the identifier point cloud accumulation database 112, thereby enriching the data of the first point cloud and the second point cloud for the same identifier, enabling more accurate position calculation.

[0031] <Vehicle position calculation unit> The vehicle position calculation unit 115 calculates a first straight line from the first point cloud, and similarly calculates a second straight line from the second point cloud. Then, based on the intersection of the calculated first and second straight lines, the relative position, which is made up of the relative distance and direction between the identifier 205 and the vehicle 107, can be calculated. At this time, the relative attitude of the identifier 205 with respect to the vehicle 107 may be calculated. Furthermore, the position information and installation orientation of the identifier 205 in the global coordinate system can be acquired from the identifier information database 114 . This makes it possible to calculate the position information and installation attitude of the vehicle 107 by combining the identifier 205 with the relative position of the vehicle 107.

[0032] <Identifier point cloud accumulation database> The identifier point cloud storage database 112 stores a first point cloud 207A and a second point cloud 207B acquired by the vehicle 107 at a first time. Furthermore, the first point cloud 206A and the second point cloud 206B acquired at the second time when matching is confirmed by the identifier point cloud matching unit 113 are stored as data of the first point cloud and the second point cloud. If the vehicle 107 is traveling, the location will also change depending on the time, so if the vehicle 107 is traveling while approaching the identifier 205, the location at the second time will be closer to the identifier than the location at the first time. When the vehicle 107 is stopped, the data of the first point cloud 206A and the second point cloud 206B are acquired from the same position as at the first time and are stored in the identifier point cloud storage database 112. The identifier point cloud accumulation database 112 may be realized by a storage device such as a hard disk drive, a semiconductor memory element such as a flash memory, or an optical disk.

[0033] <Identifier Information Database> The identifier information database 114 is a database that stores the position information and installation attitude (global coordinate system) of the identifier 205. Furthermore, the identifier information database 114 may be a distributed storage such as a cloud server, and may be shared by multiple vehicles 107 .

[0034] <Control unit> The control unit 116 is a vehicle control application that uses the estimated speed of the vehicle 107. For example, it is an automatic train operation (ATO) system that automatically controls acceleration and deceleration so that the vehicle 107 travels between predetermined points (for example, between stations) according to a scheduled time (diagram). In addition to ATO, a signal system (ATP: Automatic Train Protection) that controls braking to ensure a safe distance between vehicles may also be an application that uses the estimated speed of the vehicle 107.

[0035] Some or all of the functions of the vehicle position calculation system 200 may be realized by a CPU (Central Processing Unit) (not shown) reading and executing a program from a main storage device (software), or by hardware such as a dedicated circuit, or by a combination of software and hardware. Also, some or all of the functions of the vehicle position calculation system 200 may be realized by a computer capable of communicating with the vehicle 107.

[0036] <Example of an identifier installed around a route> Next, with reference to FIG. 4, the identifiers installed around the tracks will be described. FIG. 4 is a diagram showing identifiers installed around the tracks. The identifier 205 is composed of two planes (first and second planes) perpendicular to the nearby rail surface, and may have embedded therein location information (latitude, longitude, altitude, etc.), installation information (installation height, installation orientation, etc.), and attribute information (hereinafter referred to as "attribute information") regarding the number of lines, and is installed around the line. Note that the two planes of the identifier must not be parallel, but must intersect each other. The attribute information of the identifier 205 may be information indicating only whether the line is a single track or not (binary information of 0, 1, etc.), or may be information indicating the number of lines (1, 2).

[0037] <Example of information contained in the identifier> Next, an example of information included in the identifier 205 will be described with reference to FIG. FIG. 5 is a diagram showing an example of information included in the identifier 205. As shown in FIG. The information included in the identifier 205 includes, for example, identifier ID information 211 that uniquely identifies the identifier 205, longitude 213, latitude 215, track number information 217 regarding the number of lines, and track number information 219 that indicates the number of tracks. However, the present disclosure is not limited to this, and only the latitude and longitude information may be included in the identifier 205, and the attribute information may be stored in the identifier information database 114.

[0038] <First point cloud 206A, second point cloud 206B, and rail point cloud 206C> Next, the first point cloud 206A, the second point cloud 206B, and the rail point cloud 206C will be described with reference to FIG. FIG. 6 is a diagram showing a first point cloud 206A, a second point cloud 206B, and a rail point cloud 206C at a first time. The first point group 206A and the second point group 206B extracted by the identifier recognition unit 109 are shown in FIG. Similarly, the rail point cloud 206C recognized by the rail recognition unit 110 is shown in FIG.

[0039] <Irradiation of laser light onto identifier by sensing unit> Next, irradiation of the identifier with laser light by the sensing unit 108 will be described with reference to FIG. FIG. 7 is a diagram showing how the sensing unit 108 irradiates the identifier 205 with laser light. The sensing unit 108 irradiates the identifier 205 with laser light. At this time, a collection of points of reflected light of the laser light irradiated onto the first plane is acquired, and measurement data thereof is obtained. Similarly, a collection of points of reflected light of the laser light irradiated onto the second plane is obtained, and measurement data thereof is acquired. The identifier recognition unit 109 can extract the first point group 206A and the second point group B based on the acquired measurement data.

[0040] <Irradiation of laser light onto the rail by the sensing unit> Next, irradiation of the rail 120 with laser light by the sensing unit 108 will be described with reference to FIG. FIG. 8 is a diagram showing how the sensing unit 108 irradiates the rail 120 with laser light. The sensing unit 108 irradiates the rail 120 with laser light. At this time, a collection of points of reflected light of the laser light irradiated onto the rail 120 is acquired, and measurement data thereof is obtained. Based on the acquired measurement data, the rail recognition unit 110 can extract the rail point cloud 206C and calculate the rail surface.

[0041] <An example of a point cloud of identifiers projected onto the rail surface> Next, with reference to FIG. 9, the point cloud of the identifiers projected onto the rail surface at the second time will be described. FIG. 9 is a diagram showing a point cloud projected onto the rail surface at a second time. At the second time, when the first point cloud 206A, the second point cloud 206B, and the rail point cloud 206C extracted by the vehicle 107 are projected onto the rail surface, the result is as shown in FIG.

[0042] <Superposition of the first point cloud and the second point cloud acquired at the first time> Next, superimposition of the first point cloud 207A and the second point cloud 207B acquired at the first time will be described with reference to FIG. FIG. 10 is a diagram showing a case where a first point cloud 206A and a second point cloud 206B acquired at a second time are superimposed on a first point cloud 207A and a second point cloud 207B acquired at a first time. The first point cloud 207A and the second point cloud 207B acquired at the first time are also projected onto the rail surface as shown in FIG.

[0043] <Deriving the relative position and orientation of an identifier from an identifier point cloud> Next, with reference to FIG. 11, the intersection 210 of the first straight line 208A and the second straight line 208B will be described. FIG. 11 is a diagram showing an intersection 210 between a first straight line 208A and a second straight line 208B. The vehicle position calculation unit 115 can calculate the intersection 210 of a first straight line 208A calculated from the first point cloud registered in the identifier point cloud accumulation database 112 and a second straight line 208B calculated from the second point cloud. The straight lines can be calculated by using a known approximation method such as the least squares method on the data of the first point group and the second point group, respectively. In addition, if the first straight line and the second straight line do not intersect with each other, the midpoint between the two lines where the distance between them is shortest may be used as the intersecting point. This allows the vehicle position calculation unit 115 to measure the relative position of the identifier 205 with respect to the vehicle 107. In addition, the relative orientation of the identifier 205 with respect to the vehicle 107 can also be measured.

[0044] <Location estimation processing> Next, a flowchart showing the position estimation process will be described with reference to FIG. FIG. 12 is a flowchart showing the position estimation process executed by the sensor data processing unit 202. In addition, in FIG. 12, an example will be described in which point cloud data acquired by LiDAR is used as the data acquired by the sensing unit 108, but the present invention is not limited to this. Furthermore, the processing shown in the flowchart of FIG. 7 will be described on the assumption that it is executed at each calculation cycle of the position estimation device, but it may be executed at other cycles.

[0045] (Step S101) In step S101, the sensing unit 108 acquires measurement data.

[0046] (Step S102) In step S102, the identifier recognition unit 109 extracts a first point cloud 206A and a second point cloud 206B from the measurement data.

[0047] (Step S103) In step S103, the rail recognition unit 110 extracts the rail point cloud 206C from the measurement data and obtains the rail surface.

[0048] (Step S104) In step S104, the identifier point cloud matching unit 113 projects the first point cloud 206A and the second point cloud B onto the rail surface.

[0049] (Step S105) In step S105, the identifier point cloud matching unit 113 matches the first point cloud 206A and the second point cloud 206B with the first point cloud 207A and the second point cloud 207B acquired at the first time recorded in the identifier point cloud storage database 112.

[0050] (Step S106) In step S106, the vehicle position calculation unit 115 registers the matched first point cloud 206A and second point cloud 206B in the identifier point cloud accumulation database 112.

[0051] (Step S107) In step S107, the vehicle position calculation unit 115 calculates a first straight line 208A detected from the first point cloud and a second straight line 208B detected from the second point cloud.

[0052] (Step S108) In step S108, the vehicle position calculation unit 115 determines the intersection 210 of the first straight line 208A and the second straight line 208B, and calculates the relative position from the intersection position to the vehicle 107 and the relative attitude with respect to the vehicle 107, using the position of the determined intersection 210 as the position of the identifier 205.

[0053] (Step S109) In step S109, the vehicle position calculation unit 115 calculates the position and orientation of the vehicle 107 in the external coordinate system from the installation position and orientation of the identifier 205 recorded in the identifier information database 114 and the relative position or relative orientation of the identifier 205 with respect to the vehicle 107.

[0054] (Step S110) In step S110, the vehicle position calculation unit 115 outputs the calculated position and attitude of the vehicle 107 to the control unit 116, and the process ends.

[0055] <Point cloud acquisition of identifiers at multiple locations> Next, with reference to FIGS. 13 to 15, a case where a traveling vehicle acquires point clouds of identifiers at multiple points will be described. FIG. 13 is a diagram showing acquisition of a point cloud of identifiers at a distant location. FIG. 14 is a diagram showing acquisition of a point cloud of identifiers at an intermediate point. FIG. 15 is a diagram showing acquisition of a point cloud of identifiers at nearby points. As the vehicle 107 travels, it approaches the identifier 205 and acquires point clouds of the same identifier 205 multiple times, so the number of point clouds stored in the identifier point cloud accumulation database 112 also increases.

[0056] <Point cloud projected onto the rail surface of the identifier> Next, the point cloud of the identifier projected onto the rail surface will be described with reference to FIGS. FIG. 16 is a diagram showing a point cloud of the identifiers projected onto the rail surface at the first location. FIG. 17 is a diagram showing a point cloud of the identifiers projected onto the rail surface at the second location. The first point cloud 207A and the second point cloud 207B acquired at the first location are projected as shown in FIG. Similarly, the first point cloud 206A and the second point cloud 206B acquired at the second location are projected as shown in FIG.

[0057] <Identifier point cloud overlay> Next, superimposition of the first point cloud and the second point cloud will be described with reference to FIG. FIG. 18 is a diagram showing the first and second point clouds superimposed on each other. The first point cloud 206A and the second point cloud 206B acquired at the second location can be superimposed with the first point cloud 207A and the second point cloud 207B acquired at the first location to create point cloud data including the first point cloud 206A and the second point cloud 206B acquired at the second location.

[0058] <When the first and second planes that make up the identifier are not perpendicular to the rail surface> Next, with reference to FIGS. 19 and 20, a case where the first and second planes constituting the identifier 205 are not perpendicular to the rail surface will be described. FIG. 19 is a diagram showing a case where the first and second planes that constitute the rail are not perpendicular to the rail surface. FIG. 20 is a diagram showing a point cloud of identifiers projected onto the rail surface and stored in FIG. 19, when the first and second planes constituting the identifier 205 are not perpendicular to the rail surface, the point cloud 206 acquired by the vehicle 107 from the identifier 205 is detected as being shifted in the vehicle traveling direction, as shown in Fig. 20. This makes it difficult to estimate the vehicle position.

[0059] <Actions and Effects> The vehicle position calculation system 200 according to the first embodiment of the present disclosure has been described above. A vehicle position calculation system 200 according to a first embodiment of the present disclosure mainly includes a sensing unit 108, an identifier point cloud position calculation unit 111, an identifier point cloud matching unit 113, and a vehicle position calculation unit 115. According to the present disclosure, by deriving a first line and a second line based on data consisting of a plurality of point clouds, it is possible to obtain the first line and the second line by averaging errors at each point caused by the resolution of the sensor, and by determining a position as the intersection of the two lines, it becomes possible to calculate a position with high accuracy. Furthermore, by increasing the data of the above point cloud in a time series manner, the estimation accuracy of the first straight line 208A and the second straight line 208B can be improved, and the estimation accuracy of the intersection point 210, which is considered to be the position of the identifier 205, can be further improved. This allows the position to be calculated with an accuracy higher than the resolution of the sensor. [Example]

[0060] Next, with reference to FIG. 21, a configuration of a vehicle position calculation system 200 according to a second embodiment will be described. FIG. 21 is a diagram showing a train stopping at a station with railroad signs and platform doors. The vehicle position calculation system 200 according to the second embodiment differs from the first embodiment in that the identifier 205 is configured from two planes of existing equipment perpendicular to the rail surface. In the following description, the same or equivalent components as those in the first embodiment are denoted by the same reference numerals, and the description thereof will be simplified or omitted.

[0061] The vehicle position calculation system 200 mounted on the vehicle 107 according to the second embodiment is the same as that of the first embodiment, and therefore a description thereof will be omitted.

[0062] <identifier> The identifier 205 according to the second embodiment is composed of a railway sign 205A and platform doors 205B installed in a station, and the railway sign 205A and the platform doors 205B are installed at positions separate from each other. Furthermore, both the railroad sign 205A and the platform screen doors 205B are installed approximately perpendicular to the rail surface.

[0063] <Intersection point where extended lines intersect> Next, referring to FIG. 22, the intersection point where the first straight line 208A and the line extending the second straight line 208B intersect will be described in the case where the first and second faces of the identifier 205 are spaced apart. FIG. 22 is a diagram showing the intersection of the first straight line 208A and the second straight line in the second embodiment. In the second embodiment, the identifier recognition unit 109 can acquire the first point cloud 206A by the sensing unit 108 acquiring measurement data by regarding the railroad sign 205A as a first plane. Similarly, the sensing unit 108 can acquire measurement data by regarding the platform door 205B as a second plane, and can acquire a second point cloud 206B.

[0064] The vehicle position calculation unit 115 can calculate an intersection 210 between a first straight line 208A calculated from the first point group 206A and a second straight line 208B calculated from the second point group 206B.

[0065] At this time, the vehicle position calculation unit 115 can estimate the vehicle position and attitude based on the calculated intersection 210, similar to the first embodiment.

[0066] <Actions and Effects> The vehicle position calculation system 200 according to the second embodiment of the present disclosure has been described above. In the vehicle position calculation system 200 according to the second embodiment of the present disclosure, the vehicle position calculation unit 115 calculates a first straight line 208A and a second straight line 208B based on a first plane and a second plane that are installed at separate positions, and can calculate an intersection 210 of the first straight line 208A and the second straight line 208B. This makes it possible to measure the vehicle position and attitude with high accuracy and reliability even when the first plane and the second plane are not integrated. In other words, when calculating the vehicle position according to the present disclosure, it is possible to calculate the vehicle position by utilizing existing structures and knowing in advance the position of the intersection between the first plane and the second plane of the existing structure without setting up a dedicated identifier.

[0067] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure. For example, in the above first and second embodiments, the examples have been described in which the relative positions and the like are calculated using all of the point cloud data acquired in time series. However, it is considered that the number of point clouds projected onto the first plane and the second plane increases as the distance between the vehicle and the identifier decreases, and the accuracy of the point cloud data also improves. Therefore, for point cloud data acquired in a time series, data processing may be performed by weighting the point cloud data when the distance between the identifier and the vehicle is shorter. Furthermore, with regard to the data used for processing calculation, only data in which the number of point clouds projected onto the first plane and the second plane is equal to or greater than a certain number may be adopted. The present disclosure includes the following aspects.

[0068] (Aspect 1) A method for calculating the position of a vehicle traveling on a rail, comprising: Irradiating light onto an identifier having a first plane and a second plane that are substantially perpendicular to a rail surface that is a surface formed by the upper surface of the rail and that intersect with each other, and extracting a first point cloud and a second point cloud as a collection of reflected light from the first plane and the second plane, respectively; Calculating a relative position of the identifier with respect to the vehicle based on an intersection of a first straight line calculated from the first point cloud and a second straight line calculated from the second point cloud. A vehicle position calculation method comprising:

[0069] (Aspect 2) In the vehicle position calculation method according to aspect 1, The first point cloud and the second point cloud are a collection of points of reflected light of laser light irradiated by the LiDAR method. A vehicle position calculation method comprising:

[0070] (Aspect 3) In the vehicle position calculation method according to aspect 1 or 2, The first straight line and the second straight line each have the following characteristics: The first point cloud and the second point cloud acquired at the first time and registered in the identifier point cloud database are superimposed on the first point cloud and the second point cloud acquired at the second time, and the difference is calculated from the superimposed point clouds. A vehicle position calculation method comprising:

[0071] (Aspect 4) In the vehicle position calculation method according to aspect 3, A point cloud of the rail, which is a collection of reflected light from the top surface of the rail, is extracted, and the rail surface is calculated from the point cloud of the rail; projecting the first point cloud and the second point cloud acquired at the first time and the first point cloud and the second point cloud acquired at the second time onto the rail surface, and if they match, registering the first point cloud and the second point cloud acquired at the second time in the identifier point cloud database; A vehicle position calculation method comprising:

[0072] (Aspect 5) In the vehicle position calculation method according to any one of aspects 1 to 4, The first and second planes of the identifier are spaced apart from each other. A vehicle position calculation method comprising:

[0073] (Aspect 6) A vehicle position calculation system for a vehicle traveling on a rail, comprising: a sensing unit that irradiates light onto an identifier having a first plane and a second plane that are substantially perpendicular to the rail surface and intersect with each other; and an identifier point cloud position calculation unit that extracts a first point cloud and a second point cloud as a collection of reflected light from the first plane and the second plane, respectively; a vehicle position calculation unit that calculates a relative position of the identifier with respect to the vehicle based on an intersection of a first straight line calculated from the first point group and a second straight line calculated from the second point group; A vehicle position calculation system comprising:

[0074] (Aspect 7) In the vehicle position calculation system according to aspect 6, an identifier point cloud matching unit that matches a first point cloud and a second point cloud acquired at a first time and registered in an identifier point cloud database with a first point cloud and a second point cloud acquired at a second time; a vehicle position calculation unit that registers the first point cloud and the second point cloud acquired at the second time in the identifier point cloud database when a match is found; A vehicle position calculation system comprising: [Explanation of symbols]

[0075] 100 Automatic Train Operation System 101 Speed ​​Generator 102 Speed ​​calculation device 103 Position estimation device 104 Automatic train operation device 105 Vehicle information control device 106 Braking / driving control device 107 vehicles 108 Sensing unit 109 Identifier Recognition Unit 110 Rail Recognition Unit 111 Identifier point cloud position calculation unit 112 Identifier Point Cloud Accumulation Database 113 Identifier point cloud matching unit 114 Identifier Information Database 115 Vehicle position calculation unit 116 Control Unit 120 Rail 200 Vehicle Position Calculation System 202 Sensor data processing unit 205 Identifier 206A First point cloud at second location 206B Second point cloud at second location 207A First point cloud at first location 207B Second point cloud at first location 208A First Line 208B Second Line 211 Identifier ID Information 213 longitude 215 latitude Platform 217 information 219 Track Number Information

Claims

1. A method for calculating the position of a vehicle traveling on a rail, comprising: Irradiating light onto an identifier having a first plane and a second plane that are substantially perpendicular to the rail surface and intersect with each other; extracting a first point group and a second point group as collections of reflected light from the first plane and the second plane, respectively; Calculating a relative position of the identifier with respect to the vehicle based on an intersection of a first line calculated from the first point cloud and a second line calculated from the second point cloud. A vehicle position calculation method comprising:

2. 2. The vehicle position calculation method according to claim 1, The first point cloud and the second point cloud are a collection of points of reflected light of laser light irradiated by the LiDAR method. A vehicle position calculation method comprising:

3. 3. The vehicle position calculation method according to claim 1, The first straight line and the second straight line each have the following characteristics: The first point cloud and the second point cloud acquired at the first time and registered in the identifier point cloud database are superimposed on the first point cloud and the second point cloud acquired at the second time, and the image is calculated from the superimposed point clouds. A vehicle position calculation method comprising:

4. 4. The vehicle position calculation method according to claim 3, A point cloud of the rail, which is a collection of reflected light from the top surface of the rail, is extracted, and the rail surface is calculated from the point cloud of the rail; projecting the first point cloud and the second point cloud acquired at the first time and the first point cloud and the second point cloud acquired at the second time onto the rail surface, and if they match, registering the first point cloud and the second point cloud acquired at the second time in the identifier point cloud database; A vehicle position calculation method comprising:

5. 3. The vehicle position calculation method according to claim 1, The first and second planes of the identifier are spaced apart from each other. A vehicle position calculation method comprising:

6. A position calculation system for a vehicle traveling on a rail, comprising: a sensing unit that irradiates light onto an identifier having a first plane and a second plane that are substantially perpendicular to the rail surface and intersect with each other; an identifier point cloud position calculation unit that extracts a first point cloud and a second point cloud as collections of reflected light from the first plane and the second plane, respectively; a vehicle position calculation unit that calculates a relative position of the identifier with respect to the vehicle based on an intersection of a first straight line calculated from the first point group and a second straight line calculated from the second point group; A vehicle position calculation system comprising:

7. 7. The vehicle position calculation system according to claim 6, an identifier point cloud matching unit that matches a first point cloud and a second point cloud acquired at a first time and registered in an identifier point cloud database with a first point cloud and a second point cloud acquired at a second time; a vehicle position calculation unit that registers the first point cloud and the second point cloud acquired at the second time in the identifier point cloud database when a match is found; A vehicle position calculation system comprising:

Images