Position estimating system and method

Abstract

The objective of the present invention is to provide a technology that enables an improvement in the accuracy of position estimation of a vehicle, by estimating a travel path on the basis of various information (surrounding environment information) acquired from the surrounding environment of the vehicle.　To this end, a position estimating system according to the present invention comprises: a first position estimating unit that estimates a current location of a vehicle traveling on a travel path to obtain a first estimated current location; a travel path estimating unit that uses the surrounding environment information detected by a sensor to estimate the travel path on which the vehicle is traveling, to obtain an estimated travel path; and a position correcting unit that corrects the first estimated current location on the basis of the estimated travel path and obtains a first corrected current location. In the position correcting unit, a point on the estimated travel path closest to the first estimated current location can be used as the first corrected current location.

Description

Location estimation system and method

[0001] The present invention relates to a location estimation system and method.

[0002] To achieve obstacle detection and advanced train control, which are basic functions of automated railway operation, highly accurate estimation of the vehicle's position is necessary. A conventional method involves installing multiple ground coils on the track that record accurate absolute positions and correcting the vehicle's position each time a ground coil passes. However, ground coils are expensive to install and maintain, so position estimation using the Global Navigation Satellite System (GNSS) is being considered as an alternative method.

[0003] One GNSS-based position estimation technique known as map matching, which corrects position coordinates measured by GNSS to the nearest route, is widely used in car navigation systems, etc. However, when map matching is used to estimate the position of railway vehicles, there is a possibility that a train traveling on a down line may be incorrectly corrected to the up line due to GNSS measurement errors, and new information and processing will be required to correctly recognize the position.

[0004] Patent Document 1 describes a method for determining the nearest point on a travel route using position coordinates measured from GNSS, in which the travel route is identified using information on the track and platform number of the train in operation, and the position is corrected to be on the route.

[0005] JP 2013-99976 A

[0006] In Patent Document 1, the travel path is estimated based on specific predetermined information (e.g., a train schedule), such as the train's location on the track at a depot or station. However, in this case, building a system that acquires information from an external source, such as a traffic management system, would result in a larger system, resulting in increased system development costs and modifications. Such circumstances make it impossible to configure a position estimation system linked to the traffic management system. Furthermore, it is possible that the train schedule cannot be acquired when estimating the vehicle's position. Even if the train schedule can be acquired, the track and track number information of the train may be inaccurate because it may suddenly change due to passenger or other train disruptions or delays, making it difficult to accurately identify the vehicle's travel path. Therefore, the present invention aims to provide a technology that can improve the accuracy of vehicle position estimation by estimating the travel path based on various information acquired from the vehicle's surrounding environment (hereinafter referred to as "surrounding environment information").

[0007] In order to solve the above problem, one representative position estimation system of the present invention comprises a first position estimation unit that estimates the current position of a vehicle traveling on a traveling path to determine a first estimated current position, a traveling path estimation unit that estimates the traveling path on which the vehicle is traveling using surrounding environment information detected by a sensor to determine an estimated traveling path, and a position correction unit that corrects the first estimated current position based on the estimated traveling path to determine a first corrected current position.

[0008] According to the present invention, the driving path is estimated based on various information (surrounding environment information) acquired from the surrounding environment, such as the presence or absence of buildings beside the driving path and the positional relationship of the vehicle ahead, without relying on predetermined specific information such as a train schedule, and the current position of the vehicle is corrected based on the estimated driving path, thereby improving the accuracy of position estimation. Problems, configurations, and effects other than those described above will become clear from the description of the following embodiments.

[0009] FIG. 1 is a configuration diagram of a position estimation system in a first embodiment. FIG. 2 is a diagram showing an image of a driving path and route information. FIG. 3 is a conceptual diagram of estimating a driving path based on the presence or absence of buildings beside the driving path and the positional relationship of a vehicle ahead. FIG. 4 is a conceptual diagram of correcting the estimated current position to the estimated driving path to obtain a corrected current position. FIG. 5 is a flowchart showing the procedure for estimating and correcting the current position of a vehicle in the first embodiment. FIG. 6 is a configuration diagram of a position estimation system in a second embodiment.

[0010] In this embodiment, a method is described in which a travel path is estimated from surrounding environment information such as the presence or absence of buildings beside the travel path and the positional relationship of the vehicle ahead, and the current position of the vehicle is corrected based on the estimated travel path. In this embodiment, the travel path is defined by the line on which the vehicle is traveling and the track number of that line.

[0011] First, the configuration of the position estimation system 100 and the role of each component will be described. FIG. 1 is a configuration diagram of the position estimation system 100 in the first embodiment. The position estimation system 100 includes a first position estimation unit 101, a travel path coordinate storage unit 102, a travel path estimation unit 103, and a position correction unit 104. The first position estimation unit 101 acquires first measured position information, and the travel path estimation unit 103 acquires surrounding environment information from a vehicle 105. The position estimation system 100 may be installed either on the ground or on board, and some of the components may be installed on the ground and the rest on board. The position estimation system 100 and the vehicle 105 may be connected via wireless communication when the position estimation system 100 is on the ground, and may be connected via wireless communication or wired communication when the position estimation system 100 is on board.

[0012] (System) The first position estimation unit 101 has a function of estimating the current position of the vehicle 105 traveling on a travel path (hereinafter, the current position of the vehicle 105 estimated by the position estimation unit is referred to as the "estimated current position"). As a method for estimating the current position, a method of estimating the position using first measured position information obtained by satellite positioning such as GNSS can be mentioned. Note that the method for estimating the position is not limited to satellite positioning, and other measured position information from a sensor such as a LIDAR or a camera that measures the surrounding environment around the vehicle as two-dimensional or three-dimensional coordinate information may also be used. In this case, SLAM (Simultaneous Localization and Mapping) is used, which is a method of creating surrounding shape data at each position from a travel distance calculated from the difference between surrounding shape data of a position before movement and a current position.

[0013] The travel path coordinate storage unit 102 has a function of storing route information of travel paths measured in advance. Examples of the route information of travel paths include map information (coordinate point sequence data such as latitude and longitude at each point). FIG. 2 shows an image of a travel path and route information. FIG. 2(a) is an image of a travel path (an example in which three travel paths 1 to 3 are arranged side by side), and FIG. 2(b) shows map information for each travel path. It is assumed that the more precisely the coordinate point sequence data in the map information is measured, the higher the positional accuracy will be, and the intervals may be set according to the required positional accuracy.

[0014] The travel path estimation unit 103 has a function of estimating the travel path on which the vehicle 105 is traveling based on various information (surrounding environment information) acquired from the vehicle's surrounding environment (hereinafter, the travel path estimated by the travel path estimation unit 103 is referred to as the "estimated travel path"). The surrounding environment information is acquired from the environment inside or around the travel path, and is not particularly limited as long as it enables identification of the travel path on which the vehicle is traveling among multiple travel paths. However, information related to man-made objects that do not change much over time is preferable. For example, the presence or absence of buildings next to the travel path and the positional relationship of the vehicle to surrounding vehicles can be considered. Other possible objects include signs, signals, utility poles, overhead wires, the inner walls of tunnels, and platform doors installed on platforms next to the travel path. Furthermore, information on the estimated travel path can be, for example, information on which line and which track number on that line the vehicle is traveling on.

[0015] The sensing technology of an Advanced Driving Assistance System (ADAS) can be used to detect surrounding environment information such as the presence or absence of buildings beside the roadway and the position of the vehicle ahead. For example, image data, distance data, position data, etc. can be detected using an on-board sensor such as a camera or LIDAR that measures the surrounding environment around the vehicle in two or three dimensions.

[0016] FIG. 3 is a conceptual diagram illustrating how a driving path is estimated based on the presence or absence of buildings beside the roadway and the relative position of a vehicle ahead. As shown in FIG. 3 , for example, when traveling on two roadways, if another vehicle is present on the right side of the vehicle, the system determines that the vehicle ahead is traveling on roadway 2 on the right, and estimates that the vehicle is traveling on roadway 1 on the left. Alternatively, if a building on the left side is detected within a predetermined distance from the vehicle, the system estimates that the vehicle is traveling on roadway 1 on the left. Furthermore, more accurate roadway estimation can be achieved by combining multiple pieces of information obtained from the surrounding environment. For example, improved estimation accuracy can be expected by combining information on the presence or absence of a building on the left side within a predetermined distance with information on the presence or absence of a building on the left side within a predetermined distance, or by combining information on whether a building on the right side is located at a distance greater than a predetermined distance with information on the presence or absence of a building on the left side within a predetermined distance. Roadway estimation is performed continuously while the vehicle is traveling, and when the roadway is changed, such as when changing course, the new roadway is output.

[0017] The position correction unit 104 has a function of estimating the driving path and then correcting the point on the estimated driving path that is closest to the estimated current location as the current location of the vehicle (hereinafter, the current location on the estimated driving path obtained by correcting the estimated current location is referred to as the "corrected current location"). FIG. 4 is a conceptual diagram showing how the estimated current location is corrected on the estimated driving path to obtain the corrected current location. In this figure, the driving path estimation unit 103 estimates driving path 2 as the driving path, and then outputs a corrected current location (black square in the figure) obtained by correcting the estimated current location (white square in the figure) estimated by the first position estimation unit 101 to the closest point on driving path 2, which is the estimated driving path. The corrected current location can be displayed on a display unit (not shown) provided on the vehicle or on the ground. The estimated current location and corrected current location in the first embodiment may also be referred to as the first estimated current location and the first corrected current location, respectively.

[0018] The vehicle 105 is a vehicle that travels on a predetermined route, such as an automobile, a train, a bus, or an LRT (Light Rail Transit).

[0019] (Flowchart) Next, the procedure for estimating and correcting the current vehicle position will be described. Fig. 5 is a flowchart showing the procedure for estimating and correcting the current vehicle position in the first embodiment.

[0020] In step S11, the first position estimation unit 101 estimates the current position of the vehicle (estimated current position). In step S12, the driving path estimation unit 103 estimates the driving path from surrounding environment information such as the presence or absence of buildings beside the driving path and the positional relationship of the vehicle ahead (estimated driving path). In step S13, the position correction unit 104 acquires coordinate point sequence data of the estimated driving path from the driving path coordinate storage unit 102. In step S14, the position correction unit 104 corrects the point on the driving path that is closest to the estimated current position to be the current position of the vehicle 105 (corrected current position).

[0021] If an error occurs in the estimation of the driving path in step S12, a setting may be adopted in step S14 in which the estimated current location in step S11 is regarded as the corrected current location, or a setting may be adopted in which the estimation of the driving path (step S12) is redone.

[0022] As described above, in this embodiment, the driving path is estimated from surrounding environment information such as the presence or absence of buildings beside the driving path and the positional relationship of the vehicle ahead, and the point on the driving path closest to the estimated current position is corrected as the vehicle's current position, thereby improving the accuracy of position estimation without relying on predetermined specific information such as a train schedule. Furthermore, since the driving path is estimated based on a variety of surrounding environment information, even if one piece of information (e.g., the position of the vehicle ahead) is missing, estimation is possible using other information (e.g., the presence or absence of buildings), and accuracy can be further improved by combining multiple pieces of information (e.g., combining the position information of the vehicle ahead and information on the presence or absence of buildings). Furthermore, by selecting and discarding information useful for estimation judgment depending on the surrounding environment of the driving path, estimation accuracy can be improved using information appropriate to the environment.

[0023] Second Embodiment This embodiment differs from the first embodiment in that a second position estimation unit 206 that acquires second measured position information is added, and the estimated current position of the vehicle is corrected by a position correction unit 204. Note that the same components as those in the first embodiment are denoted by the same reference numerals, and their description will be omitted.

[0024] FIG. 6 is a configuration diagram of a position estimation system 200 according to the second embodiment.

[0025] The second position estimation unit 206 has a function of estimating a position as a device different from the first position estimation unit 101. The second position estimation unit 206 estimates a position using second measured position information different from first measured position information obtained from sensors such as an accelerometer and / or a speedometer, an IMU, and a LIDAR. When separate information processing can be performed on the first and second measured position information, the first position estimation unit 101 and the second position estimation unit 206 do not necessarily need to be different devices.

[0026] The position correction unit 204 has a function of comparing the point on the estimated travel route closest to the first estimated current location with the point on the estimated travel route closest to the second estimated current location calculated by the second position estimation unit, and discarding the information on the first and second estimated current locations if the difference is greater than a predetermined value. This reduces the probability of an error in position estimation.

[0027] As described above, when the information on the estimated current location is discarded, a setting may be adopted in which the first estimated current location is regarded as the estimated current location of the vehicle as is, or a setting may be adopted in which a message indicating that the estimated current location is unknown is output. Furthermore, after discarding, it is also possible to set the above-mentioned comparison to be repeated until the value falls below a predetermined value. Furthermore, a setting may be adopted in which, when the point on the estimated traveling path closest to the first estimated current location and the point on the estimated traveling path closest to the second estimated current location are equal to or less than a predetermined value, the former is regarded as the estimated current location of the vehicle, or a setting may be adopted in which the average point of the two is regarded as the estimated current location.

[0028] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present invention.

[0029] The following are examples of aspects that may be included in the present invention, but the present invention is not limited to these. (Aspect 1) A position estimation system comprising: a first position estimation unit that estimates the current position of a vehicle traveling on a travel path to determine a first estimated current position; a travel path estimation unit that estimates the travel path on which the vehicle is traveling using surrounding environment information detected by a sensor to determine an estimated travel path; and a position correction unit that corrects the first estimated current position based on the estimated travel path to determine a first corrected current position. (Aspect 2) The position estimation system according to Aspect 1, wherein the position correction unit determines the point on the estimated travel path that is closest to the first estimated current position as the first corrected current position. (Aspect 3) The position estimation system according to Aspect 1 or 2, wherein the surrounding environment information includes information regarding the presence or absence of buildings beside the travel path. (Aspect 4) The position estimation system according to any one of Aspects 1 to 3, wherein the surrounding environment information includes information regarding the positional relationship of the vehicle with respect to surrounding vehicles. (Aspect 5) The position estimation system according to any one of Aspects 1 to 4, wherein the position correction unit stores route information of the travel path. (Aspect 6) The position estimation system according to Aspect 5, wherein map information is used as route information for the travel path. (Aspect 7) The position estimation system according to any one of Aspects 1 to 6, wherein the estimated travel path is specified by a route and a track number on the route. (Aspect 8) The position estimation system according to any one of Aspects 1 to 7, further comprising a second position estimation unit that estimates a current position of a vehicle traveling on a travel path to determine a second estimated current position, wherein the position correction unit corrects the second estimated current position based on the estimated travel path to determine a second corrected current position, and further discards information on the first and second estimated current positions when a difference between the first corrected current position and the second corrected current position is greater than a predetermined value. (Aspect 9) The position estimation system according to Aspect 8, wherein the first position estimation unit and the second position estimation unit determine the first and second estimated current positions using different first measured position information and second measured position information, respectively.(Aspect 10) A position estimation method comprising: a step in a first position estimation unit estimating a current position of a vehicle traveling on a travel path to determine a first estimated current position; a step in a travel path estimation unit estimating a travel path on which the vehicle is traveling using surrounding environment information detected by a sensor to determine an estimated travel path; and a step in a position correction unit correcting the first estimated current position based on the estimated travel path to determine a first corrected current position.

[0030] 100, 200... Position estimation system, 101... First position estimation unit, First position estimation unit 102... Travel path coordinate storage unit, 103... Travel path estimation unit 104, 204... Position correction unit, 105... Vehicle 206... Second position estimation unit

Claims

1. A position estimation system comprising: a first position estimation unit that estimates the current position of a vehicle traveling on a driving path to obtain a first estimated current position; a driving path estimation unit that estimates the driving path on which the vehicle is traveling using surrounding environment information detected by a sensor to obtain an estimated driving path; and a position correction unit that corrects the first estimated current position based on the estimated driving path to obtain a first corrected current position.

2. A position estimation system according to claim 1, characterized in that in said position correction section, a point on said estimated travel route that is closest to said first estimated current position is set as said first corrected current position.

3. A position estimation system according to claim 1 or 2, characterized in that the surrounding environment information includes information on the presence or absence of buildings beside the roadway.

4. A position estimation system according to claim 1 or 2, characterized in that the surrounding environment information includes information regarding the positional relationship of the vehicle with respect to surrounding vehicles.

5. A position estimation system according to claim 1 or 2, further comprising a roadway information storage unit for storing route information of a road.

6. A position estimation system according to claim 5, characterized in that map information is used as the route information of the travel route.

7. A position estimation system according to claim 1 or 2, characterized in that the estimated travel route is specified by a line and a track number of the line.

8. The position estimation system of claim 1, further comprising a second position estimation unit which estimates a current position of a vehicle traveling on a travel path to obtain a second estimated current position, wherein the position correction unit corrects the second estimated current position based on the estimated travel path to obtain a second corrected current position, and further discards information on the first and second estimated current positions if a difference between the first corrected current position and the second corrected current position is greater than a predetermined value.

9. The position estimation system according to claim 8, wherein the first position estimation unit and the second position estimation unit determine the first and second estimated current locations using different first measured position information and second measured position information, respectively.

10. A position estimation method comprising: a step in a first position estimation unit of estimating a current position of a vehicle traveling on a traveling path to obtain a first estimated current position; a step in a traveling path estimation unit of estimating the traveling path on which the vehicle is traveling using surrounding environment information detected by a sensor to obtain an estimated traveling path; and a step in a position correction unit of correcting the first estimated current position based on the estimated traveling path to obtain a first corrected current position.

Images