Vehicle position information acquisition device, vehicle position information acquisition system, and vehicle position information acquisition method

Abstract

To acquire position information of an own vehicle at high speed and with high accuracy by converging types of a plurality of features in a group of data used for correction processing.SOLUTION: A feature detection unit 40 detects surrounding features from image data, that are within a predetermined range from an own vehicle and that frequently appear in any travelling environment. A reference position data group extracting unit 50 extracts reference position data group relating to shapes of the surrounding features having the same feature type as that of features detected from the image data. A comparison position data group extracting unit 20 extracts a comparison position data group relating to shapes of features having the same feature type as that of the detected surrounding features on the map. An error calculation unit 60 associates each data of the extracted reference position data group with each data of the closest comparison position data group, and calculates an error in distance between the reference position data and the comparison position data that are associated with each other. Further, a correction unit 70 corrects position information of the own vehicle based on the calculated error.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a vehicle position information acquisition device, a vehicle position information acquisition system, and a vehicle position information acquisition method. [Background technology]

[0002] In recent years, autonomous driving technology has made remarkable progress. Along with this, development is also progressing on steering assistance functions and acceleration / deceleration adjustment functions that use high-precision maps. These high-precision maps are also used to estimate vehicle position, and studies are underway to use GNSS (Global Navigation Satellite System) and GPS (Global Positioning System) combined with IMU (Inertial Measurement Unit).

[0003] However, systems that use GNSS or combine GPS with an IMU are affected by the reception conditions of the GNSS or GPS, and there is a problem in that the detection accuracy of vehicle position information deteriorates in areas with poor reception, such as inside tunnels or in urban areas with a large number of high-rise buildings.

[0004] To address this issue and compensate for the discrepancy between actual vehicle and object position information and vehicle and object position information obtained from a combination of GNSS or GPS with an IMU, a technique has been disclosed in which, for example, multiple nodes s of surrounding objects based on laser radar detection data are associated with multiple nodes m of the surrounding objects on map data, and each associated pair of nodes s and m is used to calculate errors θ, qx, qy based on a coordinate transformation formula, and the vehicle position detected by GPS is corrected based on the errors θ, qx, qy (see, for example, Patent Document 1).

[0005] In addition, a technology has been disclosed in which the road shape is extracted using a method such as converting an image in front of the vehicle to monochrome, obtaining brightness changes using a differential filter, and extracting areas with large brightness changes as white lines, and the coordinate points of the road edge on which the vehicle is traveling are extracted as the road shape from the obtained map information, and the three-dimensional road shape is estimated using the image road information and map road information, thereby eliminating image projection errors and estimating the road shape more accurately than estimating it alone (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-191239 [Patent Document 2] Japanese Patent Application Laid-Open No. 2001-331787 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the technology described in Patent Document 1 requires all data present in both the map information and the information detected by the laser radar to be used to calculate the correction value for correcting the vehicle position, which poses a problem of a large processing load when correcting the vehicle position in an extremely short period of time.

[0008] Furthermore, when considering using the technology described in Patent Document 2 to correct the vehicle position, the only type of feature used in the correction process is roads, so there is a problem that highly accurate correction process cannot be performed on roads that do not have center lines or white shoulder lines, or on roads where the white lines have partially disappeared.

[0009] Therefore, the present invention has been made in consideration of the above-mentioned problems, and aims to provide a vehicle position information acquisition device, a vehicle position information acquisition system, and a vehicle position information acquisition method that acquire vehicle position information quickly and with high accuracy by aggregating data groups to be used for correction processing from among multiple types of features. [Means for solving the problem]

[0010] Mode 1: One or more embodiments of the present invention include a position information detection unit that detects position information of the vehicle itself, a map data storage unit that stores a correspondence between the type of feature on a map and a position data group related to the shape of the feature, an imaging unit that images the foreground of the vehicle itself, a feature detection unit that detects frequently appearing surrounding features within a predetermined range from the vehicle itself from the image data of the image, a reference position data group extraction unit that extracts, from the image data of the image, position data groups related to the shapes of the surrounding features that are the same type as the detected feature as a reference position data group, and a reference position data group storage unit that stores the extracted reference position data group. and a comparison position data group extracting unit that extracts a comparison position data group from the position data group relating to the surrounding features detected by the reference position data group and the shapes of the features on the map that are the same type as the feature; a comparison position data group storage unit that stores the extracted comparison position data group; an error calculation unit that associates each piece of data of the comparison position data group stored in the reference position data group storage unit with each piece of data of the comparison position data group that is closest in distance and calculates an error in distance between the associated reference position data and the comparison position data; and a correction unit that corrects position information of the vehicle detected by the position information detection unit based on the calculated error. The error calculation unit calculates a coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data as an error in the distance between the associated reference position data and the comparison position data, and the correction unit corrects the position information of the host vehicle detected by the position information detection unit based on the coordinate transformation parameter input from the error calculation unit. We have proposed a vehicle position information acquisition device.

[0011] Form 2: One or more embodiments of the present invention propose a vehicle position information acquisition device in which the feature detection unit detects the surrounding feature that is closest to the vehicle within the predetermined range.

[0012] Mode 3: One or more embodiments of the present invention propose a vehicle position information acquisition device in which the types of features include white lines, stop lines, traffic lights, pedestrian crossings, and signs.

[0014] Mode 4: One or more embodiments of the present invention are a vehicle position information acquisition system including a vehicle position information detection device and a vehicle position information correction device, wherein the vehicle position information detection device includes a position information detection unit that detects position information of the vehicle, a map data storage unit that stores a type of feature on a map and a position data group related to the shape of the feature in association with each other, and a transmission unit that transmits the type of feature on the map, the position data group related to the shape of the feature, and the position information of the vehicle stored in the map data storage unit to the vehicle position information correction device, and the vehicle position information correction device includes one or more processors and one or more memories that are communicatively connected to the one or more processors. the one or more processors detect frequently appearing surrounding features within a predetermined range from the vehicle from image data capturing an image of the view in front of the vehicle, and store a group of position data relating to the shapes of the detected surrounding features and of the same type as the detected surrounding features from the image data as a reference position data group in a first memory, extract the group of position data relating to the shapes of the detected surrounding features and of the same type as the detected surrounding features from the map data storage unit as a comparison position data group and store it in a second memory, and associate each data of the comparison position data group stored in the second memory that is closest to each data of the reference position data group stored in the first memory, A coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data is calculated as a distance error between the associated reference position data and the comparison position data, and the position information of the vehicle detected by the position information detection unit is corrected based on the coordinate transformation parameter. We propose a vehicle position information acquisition system.

[0015] Mode 5: One or more embodiments of the present invention are a subject vehicle position information acquisition method in a subject vehicle position information acquisition device including a position information detection unit, a map data storage unit, an imaging unit, a feature detection unit, a reference position data group extraction unit, a reference position data group storage unit, a comparison position data group extraction unit, a comparison position data group storage unit, an error calculation unit, and a correction unit, the method including a first step in which the position information detection unit detects position information of the subject vehicle, a second step in which the imaging unit captures an image of a foreground of the subject vehicle, a third step in which the feature detection unit detects frequently appearing surrounding features within a predetermined range from the subject vehicle from the captured image data, and the reference position data group extraction unit extracts, from the captured image data, position data groups related to the shapes of the features that are the same type as the features detected in the third step, as a reference position data group, and corrects the extracted reference position data a fourth step of storing the comparison position data group in the reference position data group storage unit; a fifth step of the comparison position data group extraction unit extracting comparison position data groups related to the shapes of the surrounding features detected in the third step from the map data storage unit, which stores the types of the features on the map in association with position data groups related to the shapes of the features, and storing the comparison position data groups in the reference position data group storage unit; a sixth step of the error calculation unit correlating each piece of the comparison position data group stored in the reference position data group storage unit with each piece of the comparison position data group stored in the comparison position data group storage unit that is closest in distance, and calculating an error in distance between the reference position data and the comparison position data; and a seventh step of the correction unit correcting the position information of the vehicle detected in the first step based on the error calculated in the sixth step. In the sixth step, the error calculation unit calculates a coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data as an error in the distance between the associated reference position data and the comparison position data, and in the seventh step, the correction unit corrects the position information of the host vehicle detected by the position information detection unit based on the coordinate transformation parameter input from the error calculation unit. We propose a method for acquiring vehicle position information. [Effects of the Invention]

[0016] According to one or more embodiments of the present invention, by aggregating data groups to be used for correction processing from among multiple types of features, it is possible to obtain vehicle position information quickly and with high accuracy. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a diagram showing a configuration of a vehicle position information acquisition device according to an embodiment of the present invention; [Figure 2] 3 is a diagram illustrating an example of a data set obtained from map information in a vehicle position information acquisition device according to an embodiment of the present invention. FIG. [Figure 3] 1 is a diagram illustrating an example of a data set obtained from image information in a vehicle position information acquisition device according to an embodiment of the present invention. FIG. [Figure 4] 3 is a diagram illustrating an example of a map data extraction area and an image data acquisition area for a host vehicle and surrounding features according to an embodiment of the present invention. FIG. [Figure 5] FIG. 2 is a diagram showing a processing flow of the vehicle position information acquisition device according to the embodiment of the present invention. [Figure 6] 3 is a diagram illustrating an example of data input to an error calculation unit in the vehicle position information acquisition device according to the embodiment of the present invention. FIG. [Figure 7] 3 is a diagram illustrating an example of the association of data input to an error calculation unit in the vehicle position information acquisition device according to the embodiment of the present invention; FIG. [Figure 8] 5 is a diagram illustrating an example of a calculation process of a coordinate transformation parameter in an error calculation unit of the vehicle position information acquisition device according to the embodiment of the present invention. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, an embodiment of the present invention will be described with reference to FIGS.

[0019] <Embodiment> A vehicle position information acquisition device 1 according to this embodiment will be described with reference to FIGS. 1 to 8. FIG.

[0020] <Configuration of vehicle position information acquisition device 1> As shown in FIG. 1, the vehicle position information acquisition device 1 according to this embodiment includes a map data storage unit 10, a map data clipping unit 11, a coordinate conversion unit 12, a comparison position data group extraction unit 20, a comparison position data group storage unit 21, an imaging unit 30, an imaged image storage unit 31, a feature detection unit 40, a reference position data group extraction unit 50, a reference position data group storage unit 51, an error calculation unit 60, a correction unit 70, and a position information detection unit 80.

[0021] The map data storage unit 10 stores the types of features on the high-precision map in association with position data groups (point cloud information) relating to the shapes of the features. Specifically, for example, white lines WL, stop lines SL, traffic lights TL, pedestrian crossings CW, signs SN, etc. are classified as feature types, and the coordinate values ​​of these features on the high-precision map are linked and stored.

[0022] The map data cutout unit 11 cuts out map information around a specific feature detected by the feature detection unit 40 from the map data stored in the map data storage unit 10 based on the detection results of the feature detection unit 40 described below. Specifically, when the feature detection unit 40 described later detects a white line WL, in the example shown in FIG. 4, the area surrounded by the dotted line is set as the map data cut-out area, and map information of a part of the white line WL and a part of the stop line SL is cut out.

[0023] The coordinate conversion unit 12 converts the coordinates of the position data group expressed in the map coordinate system from the map information extracted by the map data extraction unit 11 into a coordinate system in which the lateral direction ahead of the vehicle is the X axis and the longitudinal direction ahead of the vehicle is the Z axis. Specifically, when the coordinate information before conversion is expressed as Equation 1 and the coordinate information after conversion is expressed as Equation 2, the coordinates are converted using the conversion formula shown in Equation 3. The database linking the position data after coordinate conversion with the type of feature is shown in FIG. 2, for example.

[0024]

number

[0025]

number

[0026]

number

[0027] The comparison position data group extraction unit 20 extracts a comparison position data group from a position data group relating to the shapes of surrounding features detected by the feature detection unit 40 (described later) and features of the same type on the map. Specifically, for example, if the feature detected by the feature detection unit 40 described later is a white line, the comparison position data group extraction unit 20 extracts the coordinate values ​​of the position information whose feature type is a white line from the database shown in Figure 2, and sets these as the comparison position data group.

[0028] The comparison position data group storage unit 21 temporarily stores the comparison position data group extracted by the comparison position data group extraction unit 20. The information of the comparison position data group stored in the comparison position data group storage unit 21 is read out by the error calculation unit 60, which will be described later, and is used in the error calculation process.

[0029] The imaging unit 30 is, for example, a stereo camera or a monocular camera, and captures a moving image of the scene in front of the host vehicle MC. The moving images captured by the imaging unit 30 are output to the feature detection unit 40, which will be described later. For example, in the example shown in Figure 4, the acquisition area of ​​image data captured by the imaging unit 30 is the area surrounded by the dashed dotted line, and image data (point cloud data) of a portion of the white line WL and a portion of the stop line SL is acquired. Here, the image data acquisition area is set wider than the map data cut-out area, taking into consideration the driving behavior of the host vehicle MC.

[0030] The captured image storage unit 31 temporarily stores the moving images captured by the imaging unit 30. More specifically, in the example shown in FIG. 4, among the moving images captured by the imaging section 30, moving image data of the image data acquisition area is temporarily stored. The moving images stored in the captured image storage unit 31 are read out by the reference position data group extraction unit 50 .

[0031] The feature detection unit 40 detects, from the image data captured by the imaging unit 30, surrounding features that frequently appear within a predetermined range from the host vehicle MC regardless of the driving environment. Examples of unfavorable driving environments include tunnels and urban areas with a large number of buildings. Examples of surrounding features that frequently appear even in such unfavorable driving environments include road shapes, particularly those containing white lines (WL). For example, in the example shown in FIG. 4, the feature detection unit 40 detects a white line WL and a pedestrian crossing CW as surrounding features within a predetermined range from the host vehicle MC. Generally, the longer the distance from the vehicle to the feature, the smaller the discrepancy between the image data and the map data. Therefore, it is preferable that the "predetermined range" be a range based on knowledge. The feature detection unit 40 may detect the surrounding feature closest to the vehicle within a predetermined range. The detected information is then output to the map data clipping unit 11 and the reference position data group extraction unit 50.

[0032] The reference position data group extraction unit 50 reads out the moving images stored in the captured image storage unit 31 from the image data captured by the imaging unit 30, and extracts a position data group relating to the shapes of surrounding features that are of the same type as the feature detected by the feature detection unit 40 as a reference position data group.

[0033] The reference position data group storage unit 51 temporarily stores the reference position data group extracted by the reference position data group extraction unit 50. The information of the reference position data group stored in the reference position data group storage unit 51 is read out by the error calculation unit 60, which will be described later, and is used in the error calculation process.

[0034] The error calculation unit 60 associates each piece of data in the reference position data group stored in the reference position data group storage unit 51 with the data in the comparison position data group stored in the comparison position data group storage unit 21 that has the closest distance, and calculates a parameter that minimizes the sum of the distances between the associated reference position data and the comparison position data as the error in the distance between the associated reference position data and the comparison position data. Specifically, for example, as shown in FIG. 6, each data of the reference position data group is associated with the data of the comparison position data group stored in the comparison position data group storage unit 21 that is closest to it, and arranged in the XZ coordinate system. Here, ◯ indicates a map point cloud, and ● indicates an image point cloud. Then, as shown in FIG. 7, the sum of the distances (L1 to L8 in FIG. 7) between the points in the map point cloud that have been associated and the corresponding points in the image point cloud is defined as a cost function as shown in Equation 4 (Step 1).

[0035]

number

[0036] Next, as shown in FIG. 8, parameters (x, z, θ) that minimize the cost function defined by Equation 4 are calculated. In the following, an example will be described in which the steepest descent method is used as a method for calculating the parameters (x, z, θ).

[0037] As the next step, Cost(x, z, θ) is partially differentiated for each of x, z, and θ using Equations 5 to 7, and the gradient at the current parameters (x, z, θ) is calculated (Step 2).

[0038]

number

[0039]

number

[0040]

number

[0041] In Equations 5 to 7, α is a learning rate indicating how much the parameters are updated each time.

[0042] Next, the parameters (x, z, θ) are updated in the direction opposite to the gradient calculated in step 2 (step 3).

[0043] Then, the processes from step 1 to step 3 are repeated until the parameters (x, z, θ) converge. Here, the converged parameters (x, z, θ) are output to the correction unit 70, which will be described later.

[0044] The correction unit 70 calculates the parameters (x, z, θ) input from the error calculation unit 60 and the parameters (map x_raw ,map z_raw ) into equation 3, and the corrected (map x_own ,map z_own )

[0045] The position information detection unit 80 detects the position information of the vehicle using a GPS or the like. The position information of the vehicle detected by the position information detection unit 80 is output to the correction unit 70.

[0046] <Processing of Vehicle Position Information Acquisition Device 1> The processing of the vehicle position information acquisition device 1 according to this embodiment will be described with reference to FIG.

[0047] The feature detection unit 40 detects, from the image data captured by the imaging unit 30, surrounding features that frequently appear within a predetermined range from the vehicle MC regardless of the driving environment (step S100). Then, the information detected by the feature detection unit 40 is output to the map data extraction unit 11.

[0048] The map data cut-out unit 11, which receives the detection information from the feature detection unit 40, cuts out map information around the specific feature detected by the feature detection unit 40 from the map data stored in the map data storage unit 10 based on the detection information from the feature detection unit 40 (step S200).

[0049] The coordinate conversion unit 12 converts the coordinates of the position data group expressed in the map coordinate system from the map information extracted by the map data extraction unit 11 into a coordinate system in which the lateral direction ahead of the vehicle is the X axis and the longitudinal direction ahead of the vehicle is the Z axis (step S300). Specifically, when the coordinate information before conversion is expressed by Equation 8 and the coordinate information after conversion is expressed by Equation 9, the coordinates are converted by the conversion formula shown in Equation 10.

[0050]

number

[0051]

number

[0052]

number

[0053] The reference position data group extraction unit 50 reads out the moving images stored in the captured image storage unit 31 from the image data captured by the imaging unit 30, and extracts a position data group relating to the shapes of surrounding features that are of the same type as the feature detected by the feature detection unit 40 as a reference position data group (step S400).

[0054] The comparison position data group extraction unit 20 extracts a comparison position data group from a position data group relating to the shapes of surrounding features detected by the feature detection unit 40 and features of the same type on the map (step S500).

[0055] The error calculation unit 60 associates each piece of data in the reference position data group stored in the reference position data group storage unit 51 with the data in the comparison position data group stored in the comparison position data group storage unit 21 that has the closest distance, and calculates a parameter that minimizes the sum of the distances between the associated reference position data and the comparison position data as the error in the distance between the associated reference position data and the comparison position data. Specifically, for example, as shown in FIG. 6, each data of the reference position data group is associated with the data of the comparison position data group stored in the comparison position data group storage unit 21 that is closest to it, and arranged in the XZ coordinate system. Here, ◯ indicates a map point cloud, and ● indicates an image point cloud. Then, as shown in FIG. 7, the sum of the distances (L1 to L8 in FIG. 7) between the points in the map point cloud that have been associated and the corresponding points in the image point cloud is defined as a cost function as shown in Equation 11 (Step 1).

[0056]

number

[0057] Next, as shown in FIG. 8, parameters (x, z, θ) that minimize the cost function defined by Equation 11 are calculated. In the following, an example will be described in which the steepest descent method is used as a method for calculating the parameters (x, z, θ).

[0058] As the next step, Cost(x, z, θ) is partially differentiated for each of x, z, and θ using Equations 12 to 14, and the gradient at the current parameters (x, z, θ) is calculated (Step 2).

[0059]

number

[0060]

number

[0061]

number

[0062] In addition, in Equations 12 to 14, α is a learning rate indicating how much the parameters are updated each time.

[0063] Next, the parameters (x, z, θ) are updated in the direction opposite to the gradient calculated in step 2 (step 3).

[0064] Then, the processes from step 1 to step 3 are repeated until the parameters (x, z, θ) converge. Here, the converged parameters (x, z, θ) are output to the correction unit 70 (step S600), which will be described later.

[0065] The correction unit 70 calculates the parameters (x, z, θ) input from the error calculation unit 60 and the parameters (map x_raw ,map z_raw ) into equation 3, and the corrected (map x_own ,map z_own ) (step S700).

[0066] <Actions and Effects> As described above, in the vehicle position information acquisition device 1 according to this embodiment, the feature detection unit 40 detects, from image data captured by the imaging unit 30, surrounding features that frequently appear within a predetermined range from the vehicle MC regardless of the driving environment. The reference position data group extraction unit 50 extracts, from the image data, a group of position data relating to the shapes of surrounding features that are the same type as the features detected by the feature detection unit 40, as a reference position data group. The comparison position data group extraction unit 20 extracts a comparison position data group from a group of position data relating to the shapes of surrounding features that are the same type on the map as the features detected by the feature detection unit 40. The error calculation unit 60 associates each piece of data in the reference position data group extracted by the reference position data group extraction unit 50 with the data in the comparison position data group extracted by the comparison position data group extraction unit that is closest in distance, and calculates the distance error between the associated reference position data and the comparison position data. Then, based on the error calculated by the error calculation unit 60, the correction unit 70 corrects the position information of the host vehicle MC detected by the position information detection unit 80 that detects the position information of the host vehicle MC. In other words, in the vehicle position information acquisition device 1 according to this embodiment, the position data on the high-precision map and the position data acquired from the image data are in a data format that corresponds to features, and the position data group on the high-precision map is matched with the image data group using the features as an index, the error calculation unit 60 further narrows down the matched data group to the data with the shortest distance and calculates the error in that distance, and the correction unit 70 corrects the position information of the vehicle MC detected by the position information detection unit 80 that detects the position information of the vehicle MC based on the error calculated by the error calculation unit 60. Therefore, highly accurate position information of the host vehicle MC can be obtained. Furthermore, in the vehicle position information acquisition device 1 according to this embodiment, a group of position data on a high-precision map is matched with a group of image data using features as indicators, and the error calculation unit 60 further narrows down the matched data group to the data with the closest distance and then calculates the error. This enables high-speed correction processing, and the position information of the vehicle MC, which changes every moment, can be obtained with high precision and speed.

[0067] Furthermore, in the vehicle position information acquisition device 1 according to this embodiment, the feature detection unit 40 detects the surrounding feature that is closest to the vehicle MC within a predetermined range. In other words, the feature detection unit 40 detects the surrounding features closest to the vehicle MC within a specified range, the reference position data group extraction unit 50 extracts from the image data a group of position data related to the shapes of surrounding features that are of the same type as the features detected by the feature detection unit 40 as a reference position data group, the comparison position data group extraction unit 20 extracts a comparison position data group from the group of position data related to the shapes of surrounding features that are of the same type on the map as the features detected by the feature detection unit 40, the error calculation unit 60 associates each data of the reference position data group extracted by the reference position data group extraction unit 50 with the data of the comparison position data group extracted by the closest comparison position data group extraction unit and calculates the error in the distance between the associated reference position data and the comparison position data, and the correction unit 70 corrects the position information of the vehicle MC detected by the position information detection unit 80 that detects the position information of the vehicle MC based on the error calculated by the error calculation unit 60. Therefore, the feature detection unit 40 further narrows down the features that can serve as indicators, so that more accurate position information of the host vehicle MC can be obtained more quickly.

[0068] In the vehicle position information acquisition device 1 according to this embodiment, the types of features include white lines, stop lines, traffic lights, pedestrian crossings, and signs. That is, the types of features are mainly white lines that indicate the shape of the road, but also include stop lines, traffic lights, crosswalks, and signs. Therefore, in areas with poor reception, such as inside tunnels, the system mainly detects white lines, and in areas with a dense urban area where high-rise buildings are lined up, such as narrow alleys and one-way streets without center lines, the system detects stop lines, traffic lights, crosswalks, signs, etc., making it possible to obtain position information of the vehicle MC with high accuracy and speed, even in places with poor reception.

[0069] In addition, in the vehicle position information acquisition device 1 according to this embodiment, the error calculation unit 60 calculates the error between the corresponding reference position data and the comparison position data, i.e., the correction value, by calculating a coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data. Therefore, highly accurate position information of the host vehicle MC can be obtained.

[0070] <Modification> In this embodiment, a vehicle position information acquisition device 1 has been described as an example, but it may also be a vehicle position information acquisition system that includes, for example, a vehicle position information detection device installed in the passenger compartment of the vehicle and a vehicle position information correction device installed on the cloud. In such a system, faster processing speeds can be expected by configuring the vehicle position information correction device, which has a large processing load, as a server.

[0071] The vehicle position information acquisition device 1 of the present invention can be realized by recording the processes of the coordinate conversion unit 12, the comparison position data group extraction unit 20, the reference position data group extraction unit 50, the error calculation unit 60, the correction unit 70, etc. on a recording medium readable by a computer system, and having the coordinate conversion unit 12, the comparison position data group extraction unit 20, the reference position data group extraction unit 50, the error calculation unit 60, the correction unit 70, etc. read and execute the programs recorded on this recording medium. The computer system here includes hardware such as an OS and peripheral devices.

[0072] Furthermore, if a WWW (World Wide Web) system is used, the "computer system" also includes the homepage provision environment (or display environment). The above program may be transmitted from a computer system that stores the program in a storage device or the like to another computer system via a transmission medium or by transmission waves in the transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has the function of transmitting information, such as a network (communication network) like the Internet or a communication line (communication line) like a telephone line.

[0073] The program may also be a program for implementing some of the above-mentioned functions, or may be a so-called differential file (differential program) that can implement the above-mentioned functions in combination with a program already stored in the computer system.

[0074] The above describes an embodiment of the present invention in detail with reference to the drawings, but the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0075] 1. Vehicle position information acquisition device 10. Map data storage unit 11: Map data extraction section 12: Coordinate conversion section 20: Comparison position data group extraction unit 21: Comparison position data group storage unit 30: Imaging unit 31: Captured image storage unit 40: Feature detection unit 50: Reference position data group extraction unit 51: Reference position data group storage unit 60;Error calculation section 70; Correction section 80: Location information detection unit

Claims

1. a position information detection unit that detects position information of the vehicle; a map data storage unit that stores the types of features on a map and position data groups related to the shapes of the features in association with each other; an imaging unit that captures an image of a view in front of the host vehicle; a feature detection unit that detects frequently appearing features within a predetermined range from the vehicle from the captured image data; a reference position data group extraction unit that extracts, from the captured image data, a position data group related to the shapes of the surrounding features that are the same type as the detected feature, as a reference position data group; a reference position data group storage unit that stores the extracted reference position data group; a comparison position data group extraction unit that extracts a comparison position data group from the position data group relating to the detected surrounding feature and the shape of the feature on the map that has the same feature type; a comparison position data group storage unit that stores the extracted comparison position data group; an error calculation unit that associates each piece of data of the reference position data group stored in the reference position data group storage unit with each piece of data of the comparison position data group stored in the comparison position data group storage unit that is closest in distance, and calculates an error in distance between the associated reference position data and the comparison position data; a correction unit that corrects the position information of the host vehicle detected by the position information detection unit based on the calculated error; Including, a correction unit correcting the position information of the vehicle detected by the position information detection unit based on the coordinate transformation parameter input from the error calculation unit; and a vehicle position information acquisition device for obtaining position information of the vehicle, the vehicle position information being corrected based on the coordinate transformation parameter input from the error calculation unit.

2. 2. The vehicle position information acquisition device according to claim 1, wherein the feature detection unit detects the surrounding feature that is closest to the vehicle within the predetermined range.

3. 3. The vehicle position information acquisition device according to claim 1, wherein the types of features include a white line, a stop line, a traffic light, a pedestrian crossing, and a sign.

4. A vehicle position information acquisition system including a vehicle position information detection device and a vehicle position information correction device, The vehicle position information detection device a position information detection unit that detects position information of the vehicle; a map data storage unit that stores the types of features on a map and position data groups relating to the shapes of the features in association with each other; a transmitter that transmits the types of features on the map stored in the map data storage unit, a group of position data relating to the shapes of the features, and the position information of the vehicle to the vehicle position information correction device; Including, The vehicle position information correction device one or more processors; and one or more memories communicatively coupled to the one or more processors; the one or more processors detect frequently appearing surrounding features within a predetermined range from the host vehicle from image data capturing an image of the view in front of the host vehicle, and store a group of position data relating to the shapes of the detected surrounding features and of the same type as the detected surrounding features from the image data as a reference position data group in a first memory; extract the group of position data relating to the shapes of the detected surrounding features and of the same type as the detected surrounding features from the map data storage unit as a comparison position data group and store them in a second memory; associate each piece of data in the reference position data group stored in the first memory with each piece of data in the comparison position data group stored in the second memory that is closest; calculate a coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data as a distance error between the associated reference position data and the comparison position data; and correct the position information of the host vehicle detected by the position information detection unit based on the coordinate transformation parameter.

5. A vehicle position information acquisition method in a vehicle position information acquisition device including a position information detection unit, a map data storage unit, an imaging unit, a feature detection unit, a reference position data group extraction unit, a reference position data group storage unit, a comparison position data group extraction unit, a comparison position data group storage unit, an error calculation unit, and a correction unit, a first step in which the position information detection unit detects position information of the vehicle; a second step in which the imaging unit captures an image of a view in front of the host vehicle; a third step in which the feature detection unit detects frequently appearing surrounding features within a predetermined range from the vehicle from the captured image data; a fourth step in which the reference position data group extraction unit extracts, from the captured image data, position data groups relating to the shapes of the features whose type is the same as that of the feature detected in the third step, as a reference position data group, and stores the extracted reference position data group in the reference position data group storage unit; a fifth step in which the comparison position data group extraction unit extracts, from the map data storage unit, which stores the types of the features on a map in association with position data groups related to the shapes of the features, comparison position data groups related to the shapes of the features and whose type is the same as that of the surrounding feature detected in the third step, and stores the extracted comparison position data groups in the reference position data group storage unit; a sixth step in which the error calculation unit associates each piece of data of the reference position data group stored in the reference position data group storage unit with each piece of data of the comparison position data group stored in the comparison position data group storage unit that is closest in distance, and calculates an error in distance between the reference position data and the comparison position data; a seventh step in which the correction unit corrects the position information of the vehicle detected in the first step based on the error calculated in the sixth step; Including, In the sixth step, the error calculation unit calculates a coordinate transformation parameter that minimizes the sum of the distances between the corresponding reference position data and the comparison position data as a distance error between the associated reference position data and the comparison position data, and in the seventh step, the correction unit corrects the position information of the vehicle detected by the position information detection unit based on the coordinate transformation parameter input from the error calculation unit.

Images