Location information calculation device and location information calculation method

Abstract

To promptly and highly accurately calculate the position of a mobile object without the need for using positioning technology such as GNSS (Global Navigation Satellite System).SOLUTION: A position information calculation device according to an embodiment of the present invention comprises: a sensor information acquisition unit acquiring sensor information which is obtained by sensing the external environment of a mobile object by a sensor; a landmark detection unit detecting a preset prescribed landmark from the sensor information and acquiring first sensor landmark information; a landmark dictionary unit preliminarily stored with second landmark information in which the landmark and position information are associated with each other; a landmark search unit acquiring the position information associated with the landmark in the first sensor landmark position information, by search using the landmark dictionary; and a mobile object position calculation unit calculating the position of the mobile object on the basis of the position information of the landmark.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] FIELD An embodiment of the present invention relates to a position information calculation device and a position information calculation method. [Background technology]

[0002] 2. Description of the Related Art For autonomous driving and autonomous running of moving bodies such as railway cars, automobiles, and robots, there has been a demand for highly accurate detection of the position of the moving body (hereinafter also referred to as "self-position").

[0003] As a technology for detecting the vehicle's own position with high accuracy, for example, there is a technology for determining the vehicle's position using the Global Navigation Satellite System (GNSS) without installing ground facilities that require maintenance. In addition, in environments where GNSS information cannot be obtained, such as tunnels and underground, a technology has been proposed for detecting the vehicle's own position by calculating the distance traveled based on the wheel diameter and the number of wheel revolutions. Another technology has been proposed for estimating the vehicle's own position by comparing images taken of the front, rear, or surroundings of the vehicle by an imaging unit mounted on the vehicle with reference images linked to location information. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-98363 A [Patent Document 2] JP 2012-208525 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the conventional technology, for example, when GNSS or autonomous navigation is used in combination to narrow down a huge number of reference images, a highly accurate positioning technology is essential. In addition, a technology that detects the vehicle's self-position by calculating the distance traveled based on the wheel diameter and the number of wheel revolutions has a problem that errors become large when the wheels are spinning. In addition, a technology that uses reference images linked to location information has a problem that there are many reference images, and it may be difficult to calculate the location information of a vehicle traveling at high speed without delay.

[0006] Therefore, the present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a position information calculation device and a position information calculation method that are capable of quickly and accurately calculating the position of a moving body without using positioning technology such as GNSS. [Means for solving the problem]

[0007] A position information calculation device according to an embodiment includes a sensor information acquisition unit that acquires sensor information obtained by sensing the external environment of a moving body using a sensor; a landmark detection unit that detects predetermined landmarks from the sensor information to acquire first sensor landmark information; a landmark dictionary unit that pre-stores second sensor landmark information in which the landmarks are linked to position information; a landmark search unit that acquires the position information linked to the landmarks in the first sensor landmark information by searching using the landmark dictionary unit; and a moving body position calculation unit that calculates the position of the moving body based on the position information of the landmarks. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an example of a schematic configuration block diagram of a railway position information calculation device according to a first embodiment. [Diagram 2] FIG. 1 is a diagram showing an example of a functional configuration block diagram of a railway position information calculation device according to a first embodiment. [Diagram 3]5 is a flowchart showing an example of a process flow of the railway position information calculation device according to the first embodiment. [Figure 4] 4 is an explanatory diagram of an example of a process for detecting a facility that is a detection target from sensor information of the railway position information calculation device according to the first embodiment; FIG. [Diagram 5] 3 is a diagram illustrating an example of a facility dictionary unit of the railway position information calculation device according to the first embodiment. FIG. [Figure 6] 4 is an explanatory diagram of an example of a process of searching for position information of each extracted area from an equipment dictionary unit of the railway position information calculation device according to the first embodiment. FIG. [Figure 7] FIG. 2 is an explanatory diagram of an example of distance information of the railway position information calculation device according to the first embodiment. [Figure 8] FIG. 11 is a diagram showing an example of a processing flow of a railway position information calculation device according to a modified example of the first embodiment. [Figure 9] FIG. 11 is a diagram showing an example of a functional configuration block diagram of a railway position information calculation device according to a second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, with reference to the drawings, an embodiment of the location information calculation device and the location information calculation method of the present invention will be described. In addition, in the drawings attached to this specification, the scale and the aspect ratio may be appropriately changed and exaggerated from those of the actual objects for the convenience of illustration and understanding. Also, when there is an arrow indicating the flow of information in the drawing, the arrow indicates the main flow of information, and information may flow through the part without the arrow.

[0010] In the following embodiment, a railroad car is taken as an example of a moving body, but is not limited thereto. The present invention can also be applied to other moving bodies, such as automobiles and robots. In addition, railroad facilities and equipment are taken as an example of a landmark for calculating the position of a moving body, but are not limited thereto. Landmarks other than railroad facilities and equipment, such as mountains and overpasses, may also be used.

[0011] (First embodiment) 1 is a schematic configuration block diagram of a first embodiment of a railway position information calculation device 1. The railway position information calculation device 1 mounted on a railway vehicle RV (hereinafter also simply referred to as a "vehicle") includes a sensor 10, a position information calculation device 20, and an interface 30.

[0012] The sensor 10 is an example of a sensor device that senses the vehicle external environment in the traveling direction or the reverse direction of the railway vehicle RV traveling on the traveling track R. The position information calculation device 20 calculates the vehicle's own position based on sensor information and distance information of the vehicle's external environment acquired by sensing by the sensor 10. The interface 30 outputs the vehicle's own position calculated by the position information calculation device 20 to the outside.

[0013] 2 is a functional configuration block diagram of the railway position information calculation device 1 of the first embodiment. The position information calculation device 20 includes a processing unit 210 and a storage unit 220. The processing unit 210 executes various information processes. The processing unit 210 includes, as functional components, a sensor information acquisition unit 20A, an equipment detection unit 20B (landmark detection unit), an equipment search unit 20C (landmark search unit), a distance information acquisition unit 20E, and a self-position calculation unit 20F. The storage unit 220 includes, for example, an equipment dictionary unit 20D (landmark dictionary unit).

[0014] The sensor information acquisition unit 20A acquires sensor information obtained by sensing the external environment of the railroad vehicle RV with the sensor 10. For example, the sensor information acquisition unit 20A acquires sensor information obtained by sensing the traveling direction or reverse direction of the railroad vehicle RV with the sensor.

[0015] The facility detection unit 20B detects predetermined facilities from the sensor information and acquires sensor facility information (first sensor landmark information).

[0016] The equipment search unit 20C acquires location information associated with equipment in the sensor equipment information by searching using the equipment dictionary unit 20D. For example, the equipment search unit 20C acquires location information associated with equipment most similar to the equipment in the sensor equipment information by searching using the equipment dictionary unit 20D.

[0017] Also, for example, when the equipment search unit 20C obtains location information linked to equipment in the sensor equipment information by searching using the equipment dictionary unit 20D, the equipment search unit 20C uses the distance information to designate as detection target candidates an area in the sensor information where the distance from the railway vehicle RV is constant.

[0018] In addition, for example, when the equipment search unit 20C acquires location information associated with equipment in the sensor equipment information by searching using the equipment dictionary unit 20D, the equipment search unit 20C searches for sensor information having location information within a predetermined range from the location information at the time immediately before the current time.

[0019] The distance information acquiring unit 20E acquires distance information of the external environment from the sensor information. That is, the distance information acquiring unit 20E acquires distance information from the vehicle to the facility included in the information sensed by the sensor 10. In addition, for example, the distance information acquiring unit 20E acquires distance information obtained by sensing the same range as the sensor information by the sensor.

[0020] The self-position calculation unit 20F calculates the position of the railroad vehicle RV based on the position information of the facility. For example, the self-position calculation unit 20F calculates the self-position of the vehicle from the position information of the facility and distance information to the facility. Also, for example, the self-position calculation unit 20F calculates the position of the railroad vehicle RV using the position information of the facility and the distance from the railroad vehicle RV to the facility calculated using the distance information.

[0021] The interface 30 is an example of a device that outputs the vehicle's own position. Here, in order to visually determine the vehicle's own position, various information may be displayed by using part or all of the sensor information of the sensor information acquisition unit 20A and the distance information of the distance information acquisition unit 20E in combination.

[0022] Some or all of the components 20A, 20B, 20C, 20E, and 20F are realized, for example, by a processor such as a CPU (Central Processing Unit) possessed by the railway vehicle RV executing software stored in a storage device possessed by the railway vehicle RV.

[0023] In addition, some or all of the units 20A, 20B, 20C, 20E, and 20F may be realized by hardware, such as a circuit board such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array), or may be realized by software executed by a processor and the hardware working together.

[0024] The storage unit 220 includes non-volatile storage media such as a Read Only Memory (ROM), a flash memory, a Solid State Drive (SSD), a Hard Disk Drive (HDD), an SD card, etc., and volatile storage media such as a Random Access Memory (RAM), a register, etc. The storage unit 220 stores various information such as programs executed by a processor included in the railcar RV.

[0025] Next, an example of the process flow of the train position information calculation device 1 according to the first embodiment will be described with reference to Fig. 3. Fig. 3 is a flowchart showing an example of the process flow of the train position information calculation device 1 according to the first embodiment.

[0026] In step S1, the sensor information acquisition unit 20A acquires sensor information sensed by the sensor 10. Examples of the acquired sensor information include a visible light image (an RGB (Red, Green, Blue) image, a grayscale image, etc.) and distance information. The visible light image is acquired, for example, from a monocular camera, a compound eye camera combining monocular cameras, or a stereo camera. Distance information is acquired from a compound eye camera, a stereo camera, a laser sensor that projects a laser and measures the light receiving position and the time the light returns to acquire position information of an object, LiDAR (Light Detection And Ranging), or the like.

[0027] In addition, machine learning methods such as ensemble learning and SVM (Support Vector Machine) that acquire distance information using visible light images as input, deep learning methods using DNN (Deep Neural Network), GNN (Graph Neural Network), GAN (Generative Adversarial Network), RNN (Recurrent Neural Network), metric learning, reinforcement learning methods, and combinations of these may also be used.

[0028] The visible light image is acquired at a predetermined resolution (VGA (Video Graphics Array), HD (High definition), Full HD, 2k, 4k, etc.), and a value expressed in a predetermined bit color (8-bit color, 24-bit color, 48-bit color, etc.) is stored for each pixel. When a visible light image is acquired using a compound eye camera, the visible light image acquired from one of the cameras may be used, or a visible light image obtained by combining both images may be used.

[0029] The distance information may be obtained from a distance image having the same resolution as the visible light image, or a value converted into a predetermined unit of length (cm, m, km, etc.) may be used as a unit expressing the distance. When distance information is obtained using a device that measures distance by projecting a laser, some pixels may not be acquired depending on the projection position. In such cases, pixels may be acquired using the acquired surrounding pixel information. For example, the average value or median value of the surrounding pixels may be used as a substitute pixel. In the first embodiment, the explanation will be given on the premise of an RGB image photographed and synthesized by a compound eye camera, and a distance image having the same resolution as the RGB image acquired by the compound eye camera.

[0030] Next, in step S2, the equipment detection unit 20B attempts to detect the equipment to be detected from the sensor information acquired in step S1, and if it is not detected (No), proceeds to step S8. If it is detected (Yes), proceeds to step S3. FIGS. 4(a) to 4(c) are explanatory diagrams of an example of a process for discovering the equipment to be detected from the acquired sensor information C1 and extracting the sensor information of the target equipment. Here, discovery refers to a process of acquiring coordinate information of the range or area of ​​the target equipment on a multidimensional plane and the name of the equipment by using sensor information as an input. Moreover, extraction refers to a process of acquiring sensor information of the range of the target equipment based on the acquired coordinate information. Furthermore, detection refers to a process of performing an extraction process after the discovery process.

[0031] The sensor information C1 is information obtained by sensing the external environment of the railroad vehicle RV in the traveling direction or the opposite direction. Here, an RGB image is used as an example, so the dimension of the sensor information is two-dimensional, and the sensor information is information on a two-dimensional plane. As an example of the detection target, a tunnel T and a sign S are set, and these target facilities are discovered using an algorithm provided in the sensor information acquisition unit 20A.

[0032] The sensor equipment coordinate information DC1 and DC2 is the coordinate information of the tunnel T and the sign S acquired according to the coordinate axes (here, the x-axis X and y-axis Y) of the smallest unit (here, pixel) that constitutes the sensor information. The format of the acquired coordinate information depends on the coordinate axes; for example, when discovering by a rectangle on a two-dimensional plane, the x-coordinate and y-coordinate of the upper left position of the rectangle and the x-coordinate and y-coordinate of the lower right position of the rectangle are acquired as the sensor equipment information.

[0033] Also, instead of the coordinates of the bottom right position of the rectangle, the length h of the height and the length w of the width from each coordinate of the top left position of the rectangle may be obtained. Also, when discovering an area using a polygon on a two-dimensional plane, the x and y coordinates may be obtained for each point constituting the polygon. Also, as an algorithm for discovering rectangles and areas, machine learning methods such as ensemble learning and SVM, deep learning methods using DNN, GNN, GAN, RNN, and metric learning, reinforcement learning methods, and methods combining these may be used. Here, the results obtained using a rectangle are shown as an example.

[0034] When detecting equipment included in the detection target, only one of the detected results may be input to the subsequent stage, or all of the detected results may be input. In that case, for example, when using DNN, if the detection reliability is in the range of 0% to 100%, the equipment whose results are to be input to the subsequent stage may be selected based on the reliability using a threshold or comparison.

[0035] In step S3, the equipment search unit 20C acquires the location information of the sensor equipment information acquired in step S2 from dictionary data in which the sensor equipment information and the location information are linked. Fig. 5 is a diagram showing an example of dictionary data (second sensor landmark information) in the equipment dictionary unit 20D. The equipment dictionary unit 20D stores the sensor equipment information and the location information linked thereto.

[0036] 6(a)-(c) are explanatory diagrams showing an example of a search process in which acquired sensor equipment information is input using dictionary data in the equipment dictionary unit 20D to acquire location information of the sensor equipment information. Here, the sensor equipment information subjected to the search may be a part of one piece of extracted sensor equipment information, or a part of them if multiple pieces of sensor equipment information are extracted. Furthermore, the location information for each extraction area obtained as a search result may be latitude and longitude information, or may be kilometer distance information.

[0037] The dictionary data stored in the equipment dictionary unit 20D may be a relational database, an object-relational database, a document-oriented database, a graph-oriented database, or the like. A deep learning method using DNN or GNN may be used for the search process. A feature point matching method between sensor equipment information may be used. For example, the sensor equipment information stored in the dictionary data may be compressed to a high dimension and stored using a deep learning method using DNN or GNN. In addition, when comparing with the input sensor equipment information, a measure in a specific distance space such as Euclidean distance, Mahalanobis distance, graph distance, or cosine similarity may be used using a deep learning method using DNN or GNN.

[0038] When searching using dictionary data stored in the equipment dictionary unit 20D, all data included in the dictionary data may be searched, or the search may be limited to sensor equipment information having location information according to a predetermined criterion. For example, when dictionary data having latitude and longitude information as location information is used, and the current time is t=1, the search range is sensor equipment information having location information within latitude 0.xx and longitude 0.yy from the location information at the previous time t=0. When dictionary data having kilometer distance information as location information is used, and the current time is t=1, the search range is sensor equipment information having location information within zz km from the location information at the previous time t=0. In this case, if no location information is obtained as a search result, the search range may be widened by expanding the reference range.

[0039] Returning to FIG. 3, in step S4, the distance information acquisition unit 20E acquires distance information sensed by the sensor 10. FIG. 7 is a diagram showing an example of distance information acquired from the sensor 10. Here, the deeper into the tunnel, the farther the distance from the vehicle, and the further away from the tunnel, the closer the distance from the vehicle. Note that the blacker the color, the farther the distance from the vehicle, and the whiter the color, the closer the distance from the vehicle. Distances are indicated in four stages, distances d1 to d4, with distance d1 representing the farthest area from the vehicle and distance d4 representing the closest area to the vehicle. Also, the entrance / exit of the tunnel T and the sign S in the figure are installed vertically, and therefore indicate that the distance from the vehicle is constant (including almost constant).

[0040] 3, in step S5, the self-position calculation unit 20F calculates the distance from the vehicle to the discovered facility using the distance information acquired in step S4. For example, when the visible light image and the distance information are configured with images of the same resolution, the same coordinate position as the coordinate position of the discovered sensor facility information is acquired from the distance information, and the distance from the vehicle to the facility is calculated using a representative value such as the average, median, or mode of the distance values ​​within that range.

[0041] In step S6, the self-position calculation unit 20F calculates the position information of the vehicle using the position information of the sensor equipment information acquired in step S3 and the distance from the vehicle to the sensor equipment information calculated in step S5. When the position information is latitude information and longitude information, the distance from the vehicle to the sensor equipment information is converted to latitude information and longitude information, and the four arithmetic operations are performed. When the position information is kilometerage information, the position of the railway vehicle RV is set as the initial position and the kilometerage of the vehicle stop position while the vehicle is stopped at the starting station of the line in advance. After departing from the starting station, the travel distance is added to the kilometerage of the vehicle stop position to calculate the vehicle position. Information on which station the vehicle is stopped at may be acquired by, for example, an input operation by the driver, or may be identified based on time information by acquiring information on the train number and operation schedule of the railway vehicle RV. Based on this, the distance from the vehicle to the sensor equipment information is converted to kilometerage, and the four arithmetic operations are performed.

[0042] In step S7, the vehicle position information calculated in step S6 is output to the external device, the interface 30. At this time, the sensor information, the sensor equipment information, and the distance information may be stored in advance, and a part or all of them may be output to the interface 30 together with the vehicle position information.

[0043] In step S8, the processing unit 210 judges whether or not an instruction to end the process of calculating the position information of the vehicle has been given. If an instruction to end the process has been given (step S8: Yes), the processing unit ends the process without acquiring new sensor information. On the other hand, if an instruction to continue the process has been given (step S8: No), the processing unit returns to step S1 and acquires new sensor information.

[0044] In this way, the railroad vehicle RV according to the first embodiment can quickly and accurately calculate the position of the moving body without using a positioning technology such as GNSS. Specifically, using sensor information obtained by sensing the traveling direction or reverse direction of the railroad vehicle RV, position information of the discovered sensor equipment information is acquired from dictionary data, and by using this together with distance information to the equipment, it becomes possible to calculate the own position of the vehicle.

[0045] In addition, in the conventional technique, a method using a reference image associated with location information has a problem that there are many reference images, and it may be difficult to calculate the location information of a vehicle traveling at high speed without delay. On the other hand, according to the method of the present embodiment, by detecting facilities, etc., processing can be simplified and high-speed and high-precision self-location calculation can be realized.

[0046] (Modification 1 of the first embodiment) In the above first embodiment, the sensor information obtained by sensing the traveling direction or reverse direction of the railroad vehicle RV is used to obtain position information of the discovered sensor equipment information from dictionary data, and the position information is used together with distance information to the equipment to calculate the vehicle's own position. Then, when the sensor equipment information is discovered from within the visible light image, the acquired sensor information is input as is, but in this modified example 1, distance information is acquired in advance and the detection range is narrowed down based on that information, thereby improving calculation efficiency.

[0047] 8 shows a flow chart for narrowing down the detection range when detecting target equipment from sensor information. Step S11 is the same as step S1 in FIG.

[0048] Next, in step S12, the distance information acquisition unit 20E acquires the distance information sensed by the sensor .

[0049] Next, in step S13, the equipment detection unit 20B attempts to detect the equipment to be detected by using the sensor information acquired in step S11 and the distance information acquired in step S12 in combination, and if the equipment cannot be detected (No), the process proceeds to step S18. If the equipment can be detected (Yes), the process proceeds to step S14.

[0050] When the sensor information and distance information are images with the same resolution, the distance value of the coordinate range of the equipment expressed in the distance information is constant, so the detection range of the sensor information is limited using that coordinate range. This makes it possible to narrow down the detection range in the sensor information using the distance information.

[0051] Furthermore, distance information with a narrowed coordinate range may be used as an algorithm for detecting target equipment from sensor information.

[0052] Steps S14 to S18 are similar to steps S3 and S5 to S8 in FIG.

[0053] (Modification 2 of the first embodiment) The self-position calculation unit 20F may calculate the self-position based on the equipment position information acquired by the equipment search unit 20C and an AI (Artificial Intelligence) learning model for self-position calculation created in advance using a reference image, without using the above-mentioned distance information.

[0054] This method requires significantly less processing time and produces highly accurate processing results compared to methods that calculate self-location based on an AI learning model without using equipment location information.

[0055] Second embodiment Next, a second embodiment will be described. The second embodiment is an example in which a position measuring unit 20G is further provided in the position information calculation device 20 included in the railway position information calculation device 1, and the self-position is calculated with higher accuracy by using the position measuring technology in combination. In the following description, the same matters as those in the first embodiment will be omitted as appropriate.

[0056] As a background, there are cases where it is desired to identify the vehicle's own position with higher accuracy depending on the vehicle in which the vehicle is installed. For example, in the case of automatic driving of vehicles such as railroad cars and automobiles where safety is the first priority, it is necessary to constantly monitor the vehicle's own position with high accuracy. In the second embodiment, an example is shown in which the position information calculation device 20 is further provided with a position measuring unit 20G to further improve the accuracy of the vehicle's own position.

[0057] Fig. 9 is a diagram showing an example of a functional configuration block diagram of a railway position information calculation device 1 (Fig. 1) according to the second embodiment. As shown in Fig. 9, in the second embodiment, the railway vehicle RV includes a sensor 10, a position information calculation device 20, and an interface 30, and the position information calculation device 20 further includes a position measurement unit 20G that measures the traveling position of the railway vehicle RV. Here, the position measurement unit 20G uses a position measurement technology such as a tachograph generator (TG) or a global navigation satellite system (GNSS).

[0058] That is, the position measuring unit 20G measures the position of the railroad vehicle RV based on a predetermined position measuring technique. Then, the self-position calculation unit 20F calculates the position of the railroad vehicle RV by performing predetermined arithmetic operations using the position information of the facility and the position information obtained by the position measuring unit 20G. This will be specifically described below.

[0059] The vehicle's position information can be acquired in multiple ways by the self-position calculation unit 20F and the position measurement unit 20G in FIG. 9. In the second embodiment, the vehicle's position information is calculated with higher accuracy by combining the respective position information by arithmetic operations or the like. The arithmetic operations may be, for example, calculation of an average value or a median value, or rejection of position information that is equal to or greater than a predetermined standard. In this case, a predetermined standard is set in advance. The predetermined standard is, for example, a case in which the vehicle's position information at the time immediately preceding the current time is latitude xx° and longitude yy°, the position information at the current time by the self-position calculation unit 20F is latitude xx+20° and longitude yy+20°, and the position information at the position measurement unit 20G is latitude xx+1° and longitude yy+1°. In addition, the predetermined standard is set to an error of within 5° latitude and 5° longitude between the position information at the current time and the position information at the time immediately preceding the current time. At this time, the position information by the self-position calculation unit 20F is rejected because it significantly deviates from the predetermined standard, and the position information by the position measurement unit 20G is set as the position information to be output to the interface 30.

[0060] The accuracy of the speed or acceleration acquired in combination in the position measuring unit 20G depends greatly on the acquisition environment. Therefore, the acceleration or speed may be calculated based on the self-position calculated at the time two before, the self-position calculated at the time one before, and each time acquired from the sensor information C1, and may be used as a correction value when calculating the self-position at the current time. For example, the time two before and the position information of the traveling vehicle are t=0, latitude x0°, and longitude y0°, the time one before and the position information are t=1, latitude x1°, and longitude y1'°, and the current time is t=2. In this case, the travel time t' and the travel distance d are obtained from the time two before and the time one before, and the respective position information, and the speed v or acceleration a is calculated from them, and the speed or acceleration a can be used as a value for correcting the speed or acceleration of the position measuring unit 20G at the current time t=2. The use as a correction value may be used every time the self-position is calculated, or the availability of use may be determined based on a predetermined criterion. Alternatively, the velocity or acceleration, or both, calculated using arithmetic operations such as the average or median may be used.

[0061] (Modification of the second embodiment) Also, for example, the self-position calculation unit 20F may calculate the distance from the railcar RV to the facility based on the position information of the facility and the position information obtained by the position measurement unit 20G, thereby correcting the distance information obtained from the distance information acquisition unit 20E. This will be specifically described below.

[0062] In the second embodiment described above, the position information calculation device 20 further includes a position measurement unit 20G, and more accurate position information is calculated by using the position measurement technology in combination with the position measurement technology. On the other hand, it is possible to correct the distance information acquired from the distance information acquisition unit 20E by using the position information of the sensor equipment information included in the sensor information acquired from the equipment search unit 20C and the position information acquired from the position measurement unit 20G. For example, the distance information from the vehicle to the detected equipment is calculated by performing arithmetic operations on the position information acquired from the equipment search unit 20C and the position information acquired from the position measurement unit 20G. By using this distance information, the distance information acquired from the distance information acquisition unit 20E can be corrected.

[0063] In addition, the speed, acceleration, or position information of the positioning technology may be used to determine the stopped state of the vehicle. In this case, the speed, acceleration, or position information is continuously measured during a certain period of time, and whether the vehicle is stopped is determined based on a predetermined criterion, so that accurate positioning is required even when the vehicle is stopped. Therefore, the stopped state of the vehicle is determined by using the speed, acceleration, or position information of the vehicle during a certain period of time shown in the first embodiment and the second embodiment. At this time, the stopped state of the vehicle may be determined by using the speed, acceleration, or position information of the positioning technology that performs the stopped state determination as a correction value, or may be determined by performing arithmetic operations with the vehicle's own position during a certain period of time.

[0064] As described above, according to the technology of each embodiment, the vehicle's position information can be calculated without using a positioning technology such as GNSS. In addition, by using the technology in combination with the positioning technology, the vehicle's position can be calculated in a composite manner, so that the vehicle's position information can be calculated with higher accuracy. Furthermore, the technology can be used to correct the speed, acceleration, or position information that is an output value used in the positioning technology, or as a fail-safe (if there are two or more position calculation methods, even if one fails, the remaining methods can be used).

[0065] The program executed in this embodiment is provided in advance in a ROM or the like, but may be provided by recording it in an installable or executable file format on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a DVD (Digital Versatile Disk). The program executed in this embodiment may be provided by storing it on a computer connected to a network such as the Internet and downloading it via the network. Furthermore, the program executed in this embodiment may be provided or distributed via a network such as the Internet.

[0066] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0067] 10...sensor, 20...position information calculation device, 20A...sensor information acquisition unit, 20B...equipment detection unit, 20C...equipment search unit, 20D...equipment dictionary unit, 20E...distance information acquisition unit, 20F...self-position calculation unit, 20G...position measurement unit, 30...interface, 210...processing unit, 220...storage unit, RV...railroad vehicle

Claims

1. a sensor information acquisition unit that acquires sensor information obtained by sensing an external environment of the moving object using a sensor; a landmark detection unit that detects a predetermined landmark from the sensor information to acquire first sensor landmark information; a landmark dictionary unit that prestores second sensor landmark information in which the landmarks are associated with position information; a landmark search unit that acquires the position information associated with the landmark in the first sensor landmark information by searching using the landmark dictionary unit; a moving object position calculation unit that calculates a position of the moving object based on the position information of the landmark; a distance information acquisition unit that acquires distance information of the external environment in the sensor information; A position measurement unit that measures the position of the moving object based on a predetermined position measurement technology, The moving object position calculation unit Calculating a position of the moving object based on the position information and the distance information of the landmark; In this case, the position information calculation device calculates the position of the moving object by performing predetermined arithmetic operations using the position information of the landmark and the position information obtained by the position positioning unit.

2. The position information calculation device according to claim 1 , wherein the moving object position calculation unit calculates the position of the moving object using the position information of the landmark and a distance from the moving object to the landmark calculated using the distance information.

3. the sensor information acquisition unit acquires the sensor information obtained by sensing a traveling direction or a reverse direction of the moving object using a sensor; The position information calculation device according to claim 1 , wherein the distance information acquisition unit acquires the distance information obtained by sensing the same range as the sensor information by a sensor.

4. The location information calculation device according to claim 1 , wherein the landmark search unit acquires the location information linked to a landmark that is most similar to the landmark in the first sensor landmark information by searching using the landmark dictionary unit.

5. 2. The location information calculation device according to claim 1, wherein when the landmark search unit acquires the location information linked to the landmark in the first sensor landmark information by searching using the landmark dictionary unit, the landmark search unit uses the distance information to designate a range in the sensor information where the distance from the moving body is constant as a detection target candidate.

6. 2. The location information calculation device according to claim 1, wherein when the landmark search unit obtains the location information linked to the landmark in the first sensor landmark information by searching using the landmark dictionary unit, the landmark search unit searches for the sensor information having location information within a predetermined range from the location information at a time immediately before the current time.

7. 2. The position information calculation device according to claim 1, wherein the moving body position calculation unit corrects the distance information acquired from the distance information acquisition unit by calculating a distance from the moving body to the landmark based on the position information of the landmark and the position information acquired by the position measurement unit.

8. a sensor information acquiring step of acquiring sensor information obtained by sensing an external environment of the moving object using a sensor; a landmark detection step of detecting a predetermined landmark from the sensor information to obtain first sensor landmark information; a landmark searching step of acquiring the location information associated with the landmark in the first sensor landmark information by searching using a landmark dictionary unit that pre-stores second sensor landmark information in which the landmark and location information are associated; a moving object position calculation step of calculating a position of the moving object based on the position information of the landmark; a distance information acquisition step of acquiring distance information of the external environment in the sensor information; A position measurement step of measuring the position of the moving object based on a predetermined position measurement technology, The moving object position calculation step includes: Calculating a position of the moving object based on the position information and the distance information of the landmark; In this case, the position information calculation method calculates the position of the moving object by performing predetermined arithmetic operations using the position information of the landmark and the position information obtained by the position positioning step.

Images