Position and orientation estimation device and position and orientation estimation method

Abstract

To provide a position and attitude estimation device and a position and attitude estimation method, which can estimate a position and an attitude of a vehicle more precisely.SOLUTION: The position and attitude estimation device of an embodiment includes a periphery measurement sensor and a position and attitude estimation unit. The periphery measurement sensor measures a distance from a vehicle and generates three-dimensional measurement data of the periphery of the vehicle. The position and attitude estimation unit estimates a position and an attitude of the vehicle on the basis of three-dimensional measurement data generated by the periphery measurement sensor and a three-dimensional periphery map stored in a storage unit.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] Embodiments according to the present invention relate to a position and orientation estimation device and a position and orientation estimation method.

Background Art

[0002] Front detection sensors (vehicle-mounted sensors) for train safe operation support and train automatic operation have been developed. When monitoring the ground around a train with a vehicle-mounted sensor, the position and orientation of the train relative to the ground constantly change during travel. Discrimination of the detection target area is greatly affected by minute changes in the position and orientation of the train relative to the ground. For appropriate operation support and automatic operation, etc., it is required to accurately estimate the position and orientation of the train relative to the ground.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] An object is to provide a position and orientation estimation device and a position and orientation estimation method capable of more accurately estimating the position and orientation of a vehicle.

Means for Solving the Problems

[0005] The position and orientation estimation device according to the present embodiment includes a peripheral measurement sensor and a position and orientation estimation unit. The peripheral measurement sensor generates three-dimensional measurement data of the periphery of the vehicle by measuring the distance from the vehicle. The position and orientation estimation unit estimates the position and orientation of the vehicle based on the three-dimensional measurement data generated by the peripheral measurement sensor and a three-dimensional surrounding map stored in the storage unit.

Brief Description of the Drawings

[0006] [Figure 1] This is a block diagram showing an example of the configuration of a railway vehicle system according to the first embodiment. [Figure 2] This figure shows an example of generating 3D measurement data from a peripheral measurement sensor according to the first embodiment. [Figure 3A] This figure shows an example of the functions of a railway vehicle system according to the first embodiment. [Figure 3B] This figure shows an example of the functions of a railway vehicle system according to the first embodiment. [Figure 4] This figure shows an example of the calculation process of the position and orientation estimation unit according to the first embodiment. [Figure 5] This figure shows an example of how the position and orientation estimation unit uses 3D measurement data according to the first embodiment. [Figure 6] This figure shows an example of how the position and orientation estimation unit uses 3D measurement data according to the first embodiment. [Figure 7] This figure shows an example of how the position and orientation estimation unit uses 3D measurement data according to the first embodiment. [Figure 8A] This figure shows an example of generating a 3D surrounding map according to the first embodiment. [Figure 8B] This figure shows an example of a three-dimensional surrounding map according to the first embodiment. [Figure 9] This is a flowchart showing an example of the operation of the position and attitude estimation device according to the first embodiment. [Figure 10] An example of 3D measurement data according to the second embodiment is shown. [Modes for carrying out the invention]

[0007] Embodiments of the present invention will be described below with reference to the drawings. These embodiments are not limiting to the present invention. The drawings are schematic or conceptual, and the proportions of each part may not necessarily be the same as those of actual objects. In the specification and drawings, elements similar to those described above with respect to previously shown drawings are denoted by the same reference numerals, and detailed explanations are omitted as appropriate.

[0008] (First Embodiment) Figure 1 is a block diagram showing an example of the configuration of a railway vehicle system according to the first embodiment.

[0009] The railway vehicle system includes a position and attitude estimation device 1 and a driving support device 2.

[0010] The position and attitude estimation device 1 estimates the position and attitude of the railway vehicle RV while it is in motion. Further details of the position and attitude estimation device 1 will be explained later.

[0011] The driver assistance device 2 provides driver assistance for the railway vehicle RV. The driver assistance device 2 provides driver assistance using the position and attitude estimated by the position and attitude estimation device 1. Driver assistance includes, for example, recognition of the driving space and detection of obstacles. Driver assistance may also include autonomous driving.

[0012] As will be explained later, the position and attitude estimation device 1 according to the first embodiment can estimate the position and attitude of the railway vehicle RV with higher accuracy. The attitude of the railway vehicle RV is, for example, its attitude relative to the ground or the track R. The attitude of the railway vehicle RV may change depending on the occupancy rate and obstacles on the track R. In addition, the attitude of the railway vehicle RV may change when the railway vehicle RV is accelerating or decelerating and when traveling on curves. By correcting the position and attitude of the railway vehicle RV with the estimated position and attitude, the driving support device 2 can provide more appropriate driving support.

[0013] Next, we will describe the details of the configuration of the position and orientation estimation device 1.

[0014] The position and orientation estimation device 1 includes a peripheral measurement sensor 10, a position measurement unit 20, a storage unit 30, a position and orientation estimation unit 40, and an output unit 50.

[0015] The peripheral measurement sensor 10 generates three-dimensional measurement data around the railway vehicle RV by measuring the distance from the railway vehicle RV. The peripheral measurement sensor 10 includes, for example, a stereo camera and a distance calculation unit. Note that the peripheral measurement sensor 10 is not limited to this, and may include, for example, a LiDAR (Light Detection And Ranging) or a ToF (Time of Flight) sensor. Also, the peripheral measurement sensor 10 may include, for example, a monocular camera. In this case, distance estimation is performed from the captured image of the monocular camera. The distance estimation is performed using, for example, SfM (Structure from Motion) or a deep learning model.

[0016] FIG. 2 is a diagram showing an example of generation of three-dimensional measurement data of the peripheral measurement sensor 10 according to the first embodiment. FIG. 2 is, for example, a captured image of a stereo camera.

[0017] The distance calculation unit of the peripheral measurement sensor 10 extracts characteristic points from the captured image as feature points FP. The distance calculation unit of the peripheral measurement sensor 10 calculates the distance of the feature point FP from the railway vehicle RV. The distance calculation unit of the peripheral measurement sensor 10 generates three-dimensional measurement data by calculating the distances of a plurality of feature points FP. The three-dimensional measurement data is, for example, three-dimensional point cloud (PC) data.

[0018] Also, the peripheral measurement sensor 10 calculates a reliability for each feature point FP along with the distance measurement.

[0019] The position measurement unit 20 measures the position of the railway vehicle RV. The position measurement unit 20 is, for example, a GNSS (Global Navigation Satellite System). Note that the position measurement unit 20 is not limited to this, and may include, for example, an IMU (Inertial Measurement Unit) or a TG (Tachogenerator).

[0020] The memory unit 30 stores a three-dimensional surrounding map. The three-dimensional surrounding map includes, for example, three-dimensional measurement data previously generated by the surrounding measurement sensor 10. More specifically, the three-dimensional surrounding map is generated by integrating three-dimensional measurement data generated at multiple different times, i.e., in a time series. The memory unit 30 may be located outside the railway vehicle RV.

[0021] The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on the 3D measurement data generated by the surrounding measurement sensor 10 and the 3D surrounding map stored in the storage unit 30. The 3D measurement data and the 3D surrounding map contain information on the position and orientation of the railway vehicle RV. By using the 3D measurement data and the 3D surrounding map, the position and orientation of the railway vehicle RV can be estimated with higher accuracy. Details of the position and orientation of the railway vehicle RV as determined by the position and orientation estimation unit 40 will be explained later with reference to Figure 4.

[0022] Furthermore, the position and orientation estimation unit 40 searches for a position on the 3D surrounding map based on the position of the railway vehicle RV measured by the position measurement unit 20. In other words, the measurement result of the position measurement unit 20 is used as the initial position when searching for a position on the 3D surrounding map.

[0023] Furthermore, the position and orientation estimation unit 40 stores the 3D measurement data generated by the surrounding measurement sensor 10 as a 3D surrounding map in the storage unit 30, based on a comparison between the 3D measurement data and the 3D surrounding map stored in the storage unit 30. More specifically, the position and orientation estimation unit 40 integrates the 3D measurement data into the 3D surrounding map if the amount of discrepancy between the 3D measurement data generated by the surrounding measurement sensor 10 and the 3D surrounding map is smaller than a predetermined amount. This allows for the sequential addition of data to the 3D surrounding map while the railway vehicle RV is in motion. By adding data multiple times at the same location, the accuracy of the 3D surrounding map can be improved.

[0024] The output unit 50 outputs the position and attitude of the railway vehicle RV estimated by the position and attitude estimation unit 40 to the outside of the position and attitude estimation device 1. More specifically, the output unit 50 outputs the position and attitude of the railway vehicle RV estimated by the position and attitude estimation unit 40 to the driving support device 2.

[0025] Furthermore, the calculation units included in the peripheral measurement sensor 10 and the position and orientation estimation unit 40 may be implemented on a single GPU (Graphics Processing Unit), or they may be implemented on separate GPUs. Alternatively, a CPU (Central Processing Unit) or the like may be used instead of a GPU.

[0026] Figure 3A shows an example of the functions of a railway vehicle system according to the first embodiment.

[0027] Based on the distance measurement results from the stereo camera and surrounding measurement sensors 10 such as LiDAR, the vehicle recognizes the driving space, identifies obstacles, and determines whether obstacles exist.

[0028] The track database shown in Figure 3A corresponds, for example, to the storage unit 30 shown in Figure 1. The track database stores track information. The position measurement shown in Figure 3A is, for example, position measurement by the position measurement unit 20. The driving assistance shown in Figure 3A is, for example, driving assistance by the driving assistance device 2.

[0029] Figure 3B shows an example of the functions of a railway vehicle system according to the first embodiment.

[0030] Figure 3B illustrates the recognition of the detection area. The detection area is a three-dimensional region defined by the vehicle clearance or building clearance of the railway vehicle RV. The detection area shown in Figure 3B represents multiple cross-sections of the three-dimensional region at multiple locations along the track R.

[0031] Conventional image recognition methods have problems such as low illumination in distant areas at night, the obscuration of the track radius by structures on curves, and difficulty in identifying the track radius in the direction of travel that needs to be monitored at track junctions.

[0032] In the example shown in Figure 3B, the system measures the train's position, corrects its position, detects the track radius (R), and recognizes the detection area. Specifically, in addition to detecting the track radius on the image, the system can narrow down the detection area by linking it with a map database. This allows for a more accurate detection area. As a result, obstacle detection can be performed with greater precision.

[0033] Next, we will describe the details of estimating the position and orientation of the railway vehicle RV.

[0034] Figure 4 shows an example of the calculation process of the position and orientation estimation unit 40 according to the first embodiment.

[0035] The lower left panel of Figure 4 shows, for example, feature point FP1 included in the 3D surrounding map. The upper left panel of Figure 4 shows, for example, feature point FP2 included in the 3D measurement data generated by the surrounding measurement sensor 10 in the current frame. In the example shown in Figure 4, the upper left panel has a different inclination than the lower left panel. This indicates that the attitude of the railway vehicle RV relative to the ground is different.

[0036] The right side of Figure 4 shows feature points FP1 and FP2. Feature point FP2 shown on the right side of Figure 4 is the starting position 101, that is, a feature point extracted from the image captured in the current frame. Feature point FP1 shown on the right side of Figure 4 is, for example, a feature point obtained at position 101 at a time earlier than the present. Trajectory 102 is the trajectory of the surrounding measurement sensor 10 (stereo camera).

[0037] The position and attitude estimation unit 40 estimates the position and attitude of the railway vehicle RV based on a comparison between the 3D point cloud data (feature point FP2) of the 3D measurement data generated by the surrounding measurement sensor 10 and the 3D point cloud data (feature point FP1) of the 3D surrounding map. The attitude of the railway vehicle RV includes, for example, the roll angle, pitch angle, and yaw angle. More specifically, the position and attitude estimation unit 40 estimates the position and attitude of the railway vehicle RV by, for example, performing calculations so that one of feature point FP1 and feature point FP2 matches the other. The amount of deviation between feature point FP1 and feature point FP2 corresponds, for example, to the tilt between the captured image in the upper left of Figure 4 and the captured image in the lower left of Figure 4. The amount of deviation between feature point FP1 and feature point FP2 can be obtained, for example, by inverse transformation and imaging.

[0038] The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on 3D measurement data generated by the surrounding measurement sensor 10, where the distance from the railway vehicle RV is less than or equal to a predetermined distance. The further away the railway vehicle RV is, for example, the lower the resolution of the captured image and the lower the accuracy of the distance at the feature point FP. By excluding feature point FPs that are far from the railway vehicle RV from the data used for estimation, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0039] Furthermore, the position and attitude estimation unit 40 may estimate the position and attitude of the railway vehicle RV based on 3D measurement data (feature points FP) from the 3D measurement data generated by the surrounding measurement sensor 10, where the reliability calculated by the surrounding measurement sensor 10 at the same time as the generation of the 3D measurement data is equal to or greater than a predetermined value.

[0040] The following describes how to further improve the accuracy of estimating the position and attitude of railway vehicle RVs. Compared to fields such as automobiles, robots, and drones, railways have, for example, long measurement distances, significant undulations around the running track (railway R), and significant changes over time due to vegetation growth. Therefore, there are cases where it is necessary to estimate the position and attitude of railway vehicle RVs with even greater accuracy.

[0041] Figure 5 shows an example of how the position and orientation estimation unit 40 utilizes 3D measurement data according to the first embodiment. Figure 5 shows, for example, an image captured by a stereo camera.

[0042] The position and orientation estimation unit 40 recognizes an object 201 that changes over time. The position and orientation estimation unit 40 recognizes the object 201 that changes over time, for example, by image recognition. The object 201 that changes over time is, for example, a tree.

[0043] The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV by excluding the 3D measurement data of the object 201 that changes over time from the 3D measurement data generated by the surrounding measurement sensor 10. If the 3D measurement data of the object 201 that changes over time is used, it may be difficult to accurately estimate the position and orientation of the railway vehicle RV. By excluding the 3D measurement data of the object 201 that changes over time, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0044] Figure 6 shows an example of how the position and orientation estimation unit 40 utilizes 3D measurement data according to the first embodiment. Figure 6 is, for example, an image captured by a stereo camera.

[0045] The position and orientation estimation unit 40 recognizes the moving object 202. The moving object 202 is, for example, another railway vehicle, such as an oncoming vehicle passing the own vehicle. If the track R is a double track, another railway vehicle may be approaching from the direction of travel of the railway vehicle RV, passing it.

[0046] The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV by excluding the 3D measurement data of the moving object 202 from the 3D measurement data generated by the surrounding measurement sensor 10. If the 3D measurement data of the moving object 202 is used, it may be difficult to accurately estimate the position and orientation of the railway vehicle RV. By excluding the 3D measurement data of the moving object 202, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0047] Figure 7 shows an example of how the position and orientation estimation unit 40 utilizes 3D measurement data according to the first embodiment. Figure 7 is, for example, an image captured by a stereo camera.

[0048] The position and orientation estimation unit 40 recognizes fixed equipment 203 installed around the running track (track R) of the railway vehicle RV. The fixed equipment 203 includes, for example, railway-specific peripheral equipment such as the track R, overhead line poles, signals, and control racks.

[0049] The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on the 3D measurement data of the fixed equipment 203, which is generated by the surrounding measurement sensor 10. By actively utilizing the 3D measurement data of the fixed equipment 203, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0050] Figure 8A shows an example of generating a 3D surrounding map according to the first embodiment. Figure 8B shows an example of a 3D surrounding map according to the first embodiment.

[0051] Figure 8A shows the image captured by the stereo camera and the result of generating 3D measurement data within the frame in the image. The result of generating 3D measurement data is shown in the upper right corner of the image. The 3D measurement data generated as shown in Figure 8A is generated at multiple different times, i.e., in a time series, and then integrated. This generates the 3D surrounding map shown in Figure 8B.

[0052] Figure 8B is a three-dimensional surrounding map obtained by a railway vehicle RV traveling in a certain direction. The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on the three-dimensional surrounding map corresponding to the direction of travel of the railway vehicle RV while it is in motion. That is, the position and orientation estimation unit 40 selects the three-dimensional surrounding map to refer to according to the direction of travel of the railway vehicle RV (for example, uphill or downhill). For example, even with the same fixed equipment 203, the appearance (extracted feature points FP) may differ depending on the viewing direction. Three-dimensional measurement data and three-dimensional surrounding maps in different directions of travel may be used for estimation, but by using three-dimensional measurement data and three-dimensional surrounding maps in the same direction of travel, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0053] Furthermore, the position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on a three-dimensional surrounding map corresponding to the conditions under which the railway vehicle RV is running. The conditions under which the vehicle is running include, for example, time of day, season, and weather, which affect the extraction of feature points FP. The three-dimensional surrounding map is stored in the storage unit 30, for example, after being tagged. By using a three-dimensional surrounding map corresponding to the conditions under which the vehicle is running, the position and orientation of the railway vehicle RV can be estimated with higher accuracy.

[0054] Next, we will explain the method for estimating position and orientation.

[0055] Figure 9 is a flowchart showing an example of the operation of the position and attitude estimation device 1 according to the first embodiment.

[0056] First, the surrounding measurement sensor 10 generates 3D measurement data by measuring the distance from the railway vehicle RV (S001). Next, the position measurement unit 20 measures the position of the railway vehicle RV as a rough estimate of its own position (S002). The position and attitude estimation unit 40 searches for a position on the 3D surrounding map (corresponding position) that corresponds to its own position (S003). Next, the position and attitude estimation unit 40 estimates the position and attitude of the railway vehicle RV based on the generated 3D measurement data and the 3D surrounding map stored in the storage unit 30 (S004).

[0057] Next, the position and orientation estimation unit 40 adds the 3D measurement data generated by the surrounding measurement sensor 10 to the 3D surrounding map (S005). However, the position and orientation estimation unit 40 refrains from adding 3D measurement data to the 3D surrounding map when the amount of displacement is large, for example.

[0058] Furthermore, the position and orientation estimation unit 40 does not necessarily have to add the 3D measurement data to the 3D surrounding map, regardless of the amount of displacement. Whether or not to add the 3D measurement data to the 3D surrounding map is set in advance.

[0059] Next, the position and orientation estimation unit 40 determines whether or not to terminate the process (S006). That is, the position and orientation estimation unit 40 determines whether or not the processing termination conditions are met. The processing termination conditions are, for example, when the railway vehicle RV is stopped at a station or when it is in a depot.

[0060] If the processing termination condition is not met (NO in step S006), step S001 is executed again. In other words, steps S001 to S005 are repeatedly executed until the processing termination determination in step S006 is made.

[0061] On the other hand, if the processing termination condition is met (NO in step S006), the process ends.

[0062] As described above, according to the first embodiment, the surrounding measurement sensor 10 generates 3D measurement data of the area around the railway vehicle RV by measuring the distance from the railway vehicle RV. The position and orientation estimation unit 40 estimates the position and orientation of the railway vehicle RV based on the 3D measurement data generated by the surrounding measurement sensor 10 and the 3D surrounding map stored in the storage unit 30. This makes it possible to estimate the position and orientation of the railway vehicle RV with higher accuracy.

[0063] Furthermore, the position measurement unit 20 is not required. In this case, the measurement result of the position measurement unit 20 is not used in step S003. When the measurement result of the position measurement unit 20 is used, the corresponding position can be searched more easily. However, the measurement result of the position measurement unit 20 is not necessarily required.

[0064] Furthermore, the 3D surrounding map does not need to be generated from the 3D measurement data; it may be pre-configured. In other words, data does not need to be added to the 3D surrounding map stored in the memory unit 30.

[0065] (Second Embodiment) Figure 10 shows an example of 3D measurement data according to the second embodiment. The second embodiment differs from the first embodiment in that 3D measurement data is generated by LiDAR.

[0066] The upper left of Figure 10 is a stereo camera image. The lower part of Figure 10 is 3D measurement data generated by LiDAR. The lower left of Figure 10 shows 3D measurement data viewed from above the railway vehicle RV, and the lower right of Figure 10 shows 3D measurement data viewed from inside the railway vehicle RV. The upper right of Figure 10 is a superimposed image of the stereo camera image and the 3D measurement data generated by LiDAR.

[0067] As in the second embodiment, 3D measurement data may be generated by LiDAR. The position and attitude estimation device 1 and position and attitude estimation method according to the second embodiment can obtain the same effects as in the first embodiment.

[0068] At least a portion of the position and attitude estimation device 1 and position and attitude estimation method according to this embodiment may be configured as hardware or as software. If configured as software, a program that implements at least some of the functions of the position and attitude estimation device 1 and position and attitude estimation method may be stored on a recording medium such as a flexible disk or CD-ROM, loaded into a computer, and executed. The recording medium is not limited to removable ones such as magnetic disks or optical disks, but may also be a fixed recording medium such as a hard disk drive or memory. Furthermore, the program that implements at least some of the functions of the position and attitude estimation device 1 and position and attitude estimation method may be distributed via a communication line such as the Internet (including wireless communication). In addition, the program may be encrypted, modulated, or compressed, and then distributed via a wired or wireless line such as the Internet, or stored on a recording medium.

[0069] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0070] 1 Position and attitude estimation device, 2 Driving support device, 10 Surrounding measurement sensor, 20 Position measurement unit, 30 Memory unit, 40 Position and attitude estimation unit, 50 Output unit, 201 Object that changes over time, 202 Moving object, 203 Fixed equipment, FP Feature point, R Track, RV Railway vehicle

Claims

1. A peripheral measurement sensor that measures the distance from a railway vehicle and generates three-dimensional measurement data of the area around the railway vehicle, A position and attitude estimation unit estimates the position and attitude of the railway vehicle based on a comparison between the three-dimensional point cloud data of the three-dimensional measurement data and the three-dimensional point cloud data of the three-dimensional surrounding map stored in the memory unit. Equipped with, The position and orientation estimation unit, (i) Recognize fixed equipment installed around the track of the railway vehicle, and estimate the position and orientation of the railway vehicle based on the three-dimensional measurement data of the fixed equipment, (ii) Among the three-dimensional measurement data, the three-dimensional measurement data is limited to those in which the distance from the railway vehicle is less than or equal to a predetermined distance, or the reliability calculated by the surrounding measurement sensor is greater than or equal to a predetermined value, (iii) Recognize objects that change over time and objects that move, and estimate the position and orientation of the railway vehicle by excluding the three-dimensional measurement data of the objects that change over time and objects that move. The aforementioned fixed equipment includes at least one of the tracks, overhead line poles, signal lights, and control racks. The aforementioned objects that change over time include trees, The aforementioned moving object is a position and orientation estimation device, including other railway vehicles passing by its own vehicle.

2. The system further includes a position measuring unit for measuring the position of the aforementioned railway vehicle, The position and attitude estimation device according to claim 1, wherein the position and attitude estimation unit searches for a position on the three-dimensional surrounding map based on the position of the railway vehicle measured by the position measurement unit.

3. The position and attitude estimation device according to claim 1, further comprising an output unit that outputs the position and attitude of the railway vehicle estimated by the position and attitude estimation unit to a driving support device that provides driving support for the railway vehicle.

4. The position and attitude estimation device according to claim 1, wherein the position and attitude estimation unit recognizes the fixed equipment and estimates the position and attitude of the railway vehicle based on the three-dimensional measurement data of the fixed equipment from the three-dimensional measurement data generated by the surrounding measurement sensor.

5. The position and attitude estimation device according to claim 1, wherein the position and attitude estimation unit estimates the position and attitude of the railway vehicle based on the three-dimensional surrounding map corresponding to the direction of travel of the railway vehicle while it is in motion.

6. The position and attitude estimation device according to claim 1, wherein the position and attitude estimation unit estimates the position and attitude of the railway vehicle based on the three-dimensional surrounding map corresponding to the conditions under which the railway vehicle is running.

7. The position and orientation estimation device according to claim 1, wherein the three-dimensional surrounding map includes the three-dimensional measurement data previously generated by the surrounding measurement sensor.

8. The position and attitude estimation device according to claim 7, wherein the position and attitude estimation unit stores the three-dimensional measurement data generated by the surrounding measurement sensor as the three-dimensional surrounding map stored in the storage unit based on a comparison between the three-dimensional measurement data and the three-dimensional surrounding map stored in the storage unit.

9. The position and orientation estimation device according to claim 1, wherein the three-dimensional surrounding map is pre-set.

10. The distance from the railway vehicle is measured and three-dimensional measurement data of the area around the railway vehicle is generated by the surrounding measurement sensor, Based on a comparison between the three-dimensional point cloud data of the three-dimensional measurement data and the three-dimensional point cloud data of the three-dimensional surrounding map stored in the memory unit, the position and orientation of the railway vehicle are estimated by the position and orientation estimation unit. It is equipped with the following: The position and orientation estimation unit, (i) Recognize fixed equipment installed around the track of the railway vehicle, and estimate the position and orientation of the railway vehicle based on the three-dimensional measurement data of the fixed equipment, (ii) Among the three-dimensional measurement data, the three-dimensional measurement data is limited to those in which the distance from the railway vehicle is less than or equal to a predetermined distance, or the reliability calculated by the surrounding measurement sensor is greater than or equal to a predetermined value, (iii) Recognize objects that change over time and objects that move, and estimate the position and orientation of the railway vehicle by excluding the three-dimensional measurement data of the objects that change over time and objects that move. The aforementioned fixed equipment includes at least one of the tracks, overhead line poles, signal lights, and control racks. The aforementioned objects that change over time include trees, The aforementioned moving object includes other railway vehicles passing by the vehicle itself, and is a method for estimating the position and orientation of the moving object.

Images