JP2019120629A - Position calculation device, position calculation program, and coordinate marker - Google Patents

Abstract

To provide a device capable of maintaining position accuracy of an automatic driving vehicle even in a place where a positioning signal from a GNSS(Global Navigation Satellite System) satellite is not receivable, and a road side system is not installed.SOLUTION: A position calculation device 100a is mounted on an automatic driving vehicle 200. The position calculation device 100a includes an on-vehicle 10 and a related position calculation part 52. The on-vehicle camera 10 photographs an image of an external environment including coordinate markers registered in DMD(Dynamic Map Data) 71 in which coordinates of arranged positions are used for automatic driving of the automatic driving vehicle 200. The related position calculation part 52 extracts identification information for identifying the coordinate marker on the basis of an image of the coordinate marker photographed by the camera device. The related position calculation part 52 acquires the coordinates of the coordinate marker from the DMD 71 on the basis of the extracted identification information. The related position calculation part 52 calculates a marker related position P(m) indicating the position of the automatic driving vehicle 200 by using the acquired coordinates of the coordinate marker.SELECTED DRAWING: Figure 7

Description

  The present invention relates to a position calculation device, a position calculation program, and a coordinate marker, which are used for calculation for specifying the position of a self-driving vehicle.

  Currently, development of autonomous vehicles is underway. In automatic driving by an autonomous car, various vehicle sensors mounted on the autonomous car detect obstacles and moving objects around the autonomous car, and the autonomous car avoids obstacles if there is an obstacle. In addition, automatic driving travel using a three-dimensional high-precision map such as dynamic map data, and a positioning correction result obtained on the order of several centimeters using positioning information transmitted by positioning satellites such as GPS satellites and quasi-zenith satellites Is being considered.

  In recent years, an infrastructure-coordinated automatic driving system has been considered for the purpose of providing road conditions and traffic information that are difficult to recognize by vehicles alone. As an example of support from the infrastructure side using the roadside system, there is a technology for distributing traveling support information from the infrastructure side to an autonomous vehicle traveling on a freeway (for example, Patent Document 1).

International Publication 2016/068273 pamphlet

  However, in locations where it is difficult to obtain positioning information from GNSS (Global Navigation Satellite System) satellites such as GPS satellites and Quasi-Zenith satellites, and where the roadside system is not installed, the position of the self-driving vehicle is It is difficult to obtain location information for identification.

  It is an object of the present invention to provide a device capable of maintaining the position accuracy for specifying the position of a self-driving vehicle even in a place where a positioning signal from a GNSS satellite can not be received and a roadside system is not installed. Do.

The position calculation device of the present invention is
A position calculation device mounted on an autonomous vehicle,
A camera device for photographing a coordinate marker registered in three-dimensional high-accuracy map data such as dynamic map data used for automatic driving of the autonomous driving vehicle, and a coordinate of the arranged position;
The identification information for identifying the coordinate marker is extracted based on the image of the coordinate marker captured by the camera device, and the coordinate of the coordinate marker is acquired from the map data based on the identification information, and the coordinates of the coordinate marker And an associated position calculation unit that calculates the position of the autonomous vehicle based on

  Since the position calculation device of the present invention can specify the position of the autonomous vehicle from the coordinate markers, it can not receive the positioning signal from the GNSS satellite, and the position accuracy of the autonomous vehicle can be obtained even at a place where the roadside system is not installed. Can maintain

FIG. 8 is a diagram of the first embodiment and a diagram for explaining a coordinate marker 90. FIG. FIG. 8 is a diagram of the first embodiment and illustrates shooting of an external environment image including a coordinate marker 90 by the on-vehicle camera 10; FIG. 9 is a diagram of the first embodiment and is a diagram showing a coordinate marker 90 a. FIG. 7 is a diagram of the first embodiment and is a diagram showing a coordinate marker 90 b. FIG. 10 is a diagram of the first embodiment and is a diagram showing a coordinate marker 90 c. FIG. 10 is a diagram of the first embodiment and is a diagram showing a coordinate marker 90 d. FIG. 2 is a diagram of the first embodiment and is a hardware configuration diagram of a position calculation device 100a. FIG. 2 is a diagram of the first embodiment and shows a data flow of the position calculation device 100a. FIG. 7 is a diagram of Embodiment 1, and a sequence diagram of the operation of the position calculation device 100 a. FIG. 2 is a diagram of the first embodiment and is a hardware configuration diagram of a position calculation device 100b. The figure of Embodiment 1 which shows the flow of data of position calculation device 100b. FIG. 8 is a diagram of the first embodiment and is a diagram showing another hardware configuration of the position calculation device 100 b.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals. In the description of the embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate.

Embodiment 1
First, an outline of the first embodiment will be described.
FIG. 1 is a diagram for explaining the coordinate marker 90. As shown in FIG.
FIG. 2 shows a shooting state of an external environment image including the coordinate marker 90 by the on-vehicle camera 10. FIG. 2 will be described in detail in the description of the operation described later.
A plurality of coordinate markers 90 that can be recognized by the on-vehicle camera 10 are disposed on the guard rail 8. The on-vehicle camera 10 captures a coordinate marker 90 by combining any one of far-infrared, near-infrared, and visible light or a plurality of on-vehicle cameras. In the first embodiment, the coordinate marker 90 is disposed on the guard rail 8, but the plurality of coordinate markers 90 are not limited to the guard rail 8, but may be arrayed in an area along a road such as "tunnel wall or sidewalk", for example. It may be done. The coordinate marker 90 can be identified by the external environment image captured by the on-vehicle camera 10.

The coordinates of each coordinate marker 90 are registered as static information in dynamic map data 71 described later. The dynamic map data 71 is stored in the auxiliary storage device 70.
In FIG. 1, the coordinates of the left coordinate marker 90 are (X1, Y1, Z1), the coordinates of the center coordinate marker 90 are (X2, Y2, Z2), and the coordinates of the right coordinate marker 90 are (X3) , Y3, Z3). The coordinates of the coordinate marker 90 are used as absolute coordinates. After recognizing the coordinate marker 90 drawn on the guardrail 8, the position calculation device 100a of the autonomous driving vehicle 200 uses the coordinate marker 90 and the lane link 7 to determine the relative position between the autonomous driving vehicle 200 and the coordinate marker 90. Identify. This relative position is the vector 6 described later. The identification of the relative position will be described later in the description of the operation.

  3 to 6 show a coordinate marker 90. FIG. FIG. 3 shows a coordinate marker 90a. FIG. 4 shows a coordinate marker 90b. FIG. 5 shows a coordinate marker 90c. FIG. 6 shows a coordinate marker 90d. The coordinate markers 90a and 90b consist only of marks. Information such as a two-dimensional barcode 3 or a numerical value 4 may be added to the coordinate marker 90. The coordinate marker 90c indicates a state in which the two-dimensional barcode 3 is attached. The coordinate marker 90d indicates a state in which a numerical value 4 indicated as AA11 is attached. The two-dimensional barcode 3 and the numerical value 4 can have identification information 91 for identifying the coordinate marker 90.

*** Description of the configuration ***
FIG. 7 is a hardware configuration diagram of the position calculation device 100a. The configuration of the position calculation device 100a will be described with reference to FIG. The position calculation device 100 a is mounted on an autonomous vehicle 200. The position calculation device 100a is a computer including hardware such as an on-vehicle camera 10, a positioning reception unit 20, an inertial measurement device 30, an input / output interface 40, a processor 50, a memory 60, and an auxiliary storage device 70. These pieces of hardware are connected to each other via a signal line 81.
The inertial measurement device 30 is hereinafter referred to as the IMU 30.

As shown in FIG. 2, the on-vehicle camera 10 captures front and side external environment images including the guard rails 8. Data of the mounting position and the mounting posture of the on-vehicle camera 10 with respect to the automatic driving vehicle 200 is stored in the auxiliary storage device 70. The auxiliary storage device 70 stores dynamic map data 71, which is map data 901 used for automatic driving of the automatic driving vehicle 200. The dynamic map data 71 is hereinafter referred to as the DMD 71. The DMD 71 includes static information, quasi-static information, quasi-dynamic information, and dynamic information. The static information of the DMD 71 is three-dimensional highly accurate basic map data, includes road surface information, lane information, three-dimensional structures, etc., and is composed of three-dimensional position coordinates representing features and linear vector data Ru. The quasi-static information, the quasi-dynamic information, and the dynamic information are dynamic data that changes from moment to moment, and are data to be superimposed on the static information based on the position reference base. The quasi-static information includes traffic control information, road construction information, wide-area weather information, etc. The semi-dynamic information includes accident information, traffic congestion information, narrow-area weather information, etc. The dynamic information is ITS information And signal information etc.).
The on-vehicle camera 10 is a camera device 902 that captures a coordinate marker 90. As described above, the coordinate of the position where the coordinate marker 90 is arranged is registered in the static information of the DMD 71.

  The positioning reception unit 20 receives a positioning signal transmitted by a GNSS satellite which is a positioning satellite, reproduces the positioning signal, and outputs it as observation information.

  The IMU 30 includes an acceleration sensor and a gyro. The IMU 30 generates and outputs observation information from the signals of the acceleration sensor and the gyro.

  The input / output interface 40 is an interface device that connects the on-vehicle camera 10, the positioning reception unit 20, the IMU 30, and the processor 50.

  The processor 50 is an integrated circuit (IC) that performs arithmetic processing, and controls other hardware. For example, the processor 50 is a central processing unit (CPU), a digital signal processor (DSP), or a graphics processing unit (GPU).

  The memory 60 is a volatile storage device. The memory 60 is also referred to as main storage or main memory. For example, the memory 60 is a random access memory (RAM). The data stored in the memory 60 is stored in the auxiliary storage device 70 as needed.

  The auxiliary storage device 70 is a non-volatile storage device. For example, the auxiliary storage device 70 is a read only memory (ROM), a hard disk drive (HDD), or a flash memory. The data stored in the auxiliary storage device 70 is loaded into the memory 60 as needed.

  The position calculation device 100 a includes functional elements of an image processing unit 51, an associated position calculation unit 52, and a position calculation unit 53. The related position calculation unit 52 includes a marker specifying unit 52a and a position correction unit 52b. The functions of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 are realized by software.

The auxiliary storage device 70 stores a position calculation program for realizing the functions of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53. The position calculation program is loaded into the memory 60 and executed by the processor 50.
Also, the DMD 71 is loaded into the memory 60 and read by the processor 50.
Further, the auxiliary storage device 70 stores an OS (Operating System). At least a portion of the OS is loaded into memory 60 and executed by processor 50. The processor 50 executes the position calculation program while executing the OS. Data obtained by executing the position calculation program is stored in a storage device such as the memory 60, the auxiliary storage device 70, a register in the processor 50 or a cache memory in the processor 50.

  The position calculation device 100 a may include a plurality of processors that replace the processor 50. The multiple processors share the role of the processor 50.

  The position calculation program can be computer-readably recorded on a non-volatile recording medium such as an optical disk or a flash memory.

*** Description of operation ***
The operation of the position calculation device 100 a will be described with reference to FIGS. 2, 8 and 9.
FIG. 8 shows the flow of data of the position calculation device 100a.
FIG. 9 shows a sequence of operations of the position calculation device 100a.

  In step S11, the positioning reception unit 20 receives the positioning signal transmitted by the GNSS satellite, reproduces the positioning signal, and outputs it as observation information. Also, the IMU 30 generates observation information from the signals of the acceleration sensor and the gyro and outputs it.

  In step S12, the position calculation unit 53 calculates the self position of the autonomous driving vehicle 200 using the positioning signal transmitted by the positioning satellite. At that time, the position calculation unit 53 uses observation information of the signal of the IMU 30 in addition to observation information of the positioning signal. The position calculation unit 53 calculates a self position by combined positioning calculation in which satellite positioning and navigation positioning are tightly coupled by applying a Kalman filter to observation information of the positioning signal and observation information of the IMU 30. The self position means the position of the autonomous driving vehicle 200.

In step S13, the on-vehicle camera 10 captures the front and side external environments including the guard rails 8 as image information as shown in FIG.
In step S14, the image processing unit 51 performs image processing such as sharpening processing and clustering processing of an external environment image. The image processing unit 51 outputs an image-processed external environment image. The external environment image output from the image processing unit 51 includes a guard rail 8 and a coordinate marker 90.

In step S15, the marker specifying unit 52a of the related position calculation unit 52 extracts the image of the coordinate marker 90 from the external environment image output from the image processing unit 51.
In step S16, the marker identifying unit 52a extracts the identification information 91 from the image of the coordinate marker 90. The identification information 91 of the coordinate marker 90 is information for identifying the coordinate marker 90. The identification information 91 is uniquely assigned to each coordinate marker 90 of the plurality of coordinate markers 90, and the coordinate marker 90 is uniquely identified by the identification information 91.
Extraction of the identification information 91 of the coordinate marker 90 is as follows.

<Type 1>
As the type 1, in the auxiliary storage device 70, correspondence information 72 in which correspondences between images of the plurality of coordinate markers 90 and identification information 91 attached to the images are registered is stored. The marker specifying unit 52a extracts the identification information 91 of the coordinate marker 90 by searching the correspondence information 72 using the extracted image of the coordinate marker 90 as a search key.

<Type 2>
In addition to the mark of the coordinate marker 90, the identification information 91 of the coordinate marker 90 is also arrange | positioned in the guardrail 8 as Type 2 as FIG.5 and FIG.6. In the case of type 2, the external environment image output from the image processing unit 51 also includes identification information 91. The marker specifying unit 52 a extracts the identification information 91 by reading the identification information 91 included in the external environment image output from the image processing unit 51.

  Thus, in step S16, the marker specifying unit 52a of the related position calculating unit 52 extracts the identification information 91 for identifying the coordinate marker 90 based on the image of the coordinate marker 90 captured by the on-vehicle camera 10.

In step S17, the marker specifying unit 52a refers to the DMD 71 and acquires the coordinates of the coordinate marker 90 from the DMD 71 based on the identification information 91 of the coordinate marker 90.
Specifically, it is as follows.
The coordinates of each coordinate marker 90 of the plurality of coordinate markers 90 are registered in the DMD 71 in association with the identification information 91. The marker specifying unit 52a searches the DMD 71 using the identification information 91 extracted in step S16 as a search key, and acquires the coordinates of the coordinate marker 90 corresponding to the identification information 91. Thus, in step S17, the marker specifying unit 52a of the related position calculation unit 52 acquires the coordinates of the coordinate marker 90 from the DMD 71 based on the identification information 91.

In step S18, the marker specifying unit 52a calculates the azimuth θ in the direction in which the coordinate marker 90 is present, based on the image of the coordinate marker 90 specified in step S15.
A method of calculating the azimuth angle θ will be described with reference to FIG.
(1) The mounting position and mounting posture of the on-vehicle camera 10 to the automatic driving vehicle 200 are known. The mounting position is coordinates (X, Y, Z) with respect to the reference position of the automatic driving vehicle 200, and the mounting posture is (roll angle, pitch angle, yaw angle) with respect to the reference posture of the automatic driving vehicle 200. The attachment position and the attachment attitude are stored in the auxiliary storage device 70.
(2) The autonomous driving vehicle 200 is operating automatically.
In the case of automatic driving, the autonomous driving vehicle 200 travels on a track corresponding to the lane link 7.
The lane link 7 is data registered in the DMD 71 and data defined in the DMD 71 as a center line of a roadway. The marker specifying unit 52 a can recognize the lane link 7 by referring to the DMD 71.
(3) The marker specifying unit 52a is a coordinate marker in the image of the coordinate marker 90 based on the image of the coordinate marker 90 extracted in step S15, the mounting position and mounting posture of the on-vehicle camera 10, and the direction of the lane link 7. An azimuthal angle of 90 can be calculated. The direction of the lane link 7 is the traveling direction of the autonomous driving vehicle 200.

A method of calculating the distance L and the marker related position P (m), which is calculated in relation to the coordinates of the coordinate marker 90, will be described with reference to FIG. The marker related position P (m) is the position of the autonomous driving vehicle 200 obtained from the coordinate marker 90 captured by the on-vehicle camera 10.
In step S19, the position correction unit 52b of the related position calculation unit 52 calculates L from the following Expression 1 and Expression 2 based on the calculated azimuth θ calculated by the marker specifying unit 52a and the coordinates of the specified coordinate marker. And calculate the marker related position P (m).
An XYZ coordinate system 5 is shown in FIG.
L = D / tan θ (equation 1)
P (m) = (Xm, Ym, Zm) + (-D, -L, Zr) (Formula 2)
(Xm, Ym, Zm) is the acquired coordinates of the coordinate marker 90.
(-D, -L, Zr) is a relative distance between the coordinate marker 90 and the autonomous driving vehicle 200.
(-D, -L, Zr) is a component of the vector 6 in FIG.
D and L are distances shown in FIG. 2 and specifically as follows.
(1) D is the shortest distance between the lane link 7 and the coordinate marker 90 in the XY plane of the XYZ coordinate system 5.
(2) L is the distance in the direction of the lane link 7 between the coordinate marker 90 and the position of the on-vehicle camera 10 in the XY plane of the XYZ coordinate system 5.
(3) Zr is the difference between the Z value of the mounting coordinates of the on-vehicle camera 10 and Zm which is the Z coordinate of the coordinate marker 90.
(4) The vector 6 is a vector directed from the coordinates of the coordinate marker 90 to the mounting position of the on-vehicle camera 10.
In the case of Formula 2, the marker related position P (m) of the autonomous driving vehicle 200 is the position of the in-vehicle camera 10, but the reference position of the autonomous driving vehicle 200 and the mounting position of the in-vehicle camera 10 are known. The reference position of the car 200 can be converted to the self position.

  As described above, using the coordinates of the coordinate marker 90, the position correction unit 52b of the related position calculation unit 52 calculates a marker related position P (m) indicating the position of the autonomous driving vehicle 200. When calculating the marker related position P (m), the related position calculation unit 52 determines the azimuth angle θ of the coordinate marker 90 with respect to the reference direction from the image of the coordinate marker 90, and uses the determined azimuth angle θ to perform automatic driving A marker related position indicating the position of the car 200 is calculated. Here, the reference direction is a direction along the lane link 7. Since the autonomous driving vehicle 200 is in an autonomous operation, as shown in FIG. 2, the direction along the lane link 7 is also the traveling direction of the autonomous driving vehicle 200.

  In step S20, the position correction unit 52b corrects the self position based on the marker related position P (m) calculated by Expression 2 and the self position determined by the position calculation unit 53. The corrected self position is used for automatic driving of the autonomous vehicle 200. Thus, the related position calculation unit 52 corrects the position of the autonomous driving vehicle 200 calculated by the position calculation unit 53 using the positioning signal, using the marker related position P (m).

If the positioning reception unit 20 can not receive the positioning signal of the GNSS satellite because the automatic driving vehicle 200 is located in a tunnel or a mountain forest, the position calculating unit 53 detects the position of the automatic driving vehicle 200 based on the observation information of the IMU 30. Determine the self position used for automatic driving. In this case, the position correction unit 52 b frequently uses the coordinate marker 90 of the guard rail 8 to correct the self position so that the calculation accuracy of the self position does not deteriorate due to the accumulation of the bias error of the IMU 30.
That is, the position calculation unit 53 calculates the position of the autonomous driving vehicle 200 based on the observation information obtained by the IMU 30. Then, the position correction unit 52b corrects the own position of the autonomous driving vehicle 200 calculated by the position calculation unit 53 using the observation information of the IMU 30 using the marker related position P (m).

*** Description of the effects of the first embodiment ***
According to the first embodiment, since the position of the automatic driving vehicle 200 can be obtained from the coordinate marker 90, automatic positioning can not be received from the GNSS satellites, and even in areas where the roadside system is not installed, the automatic operation can be performed. The position accuracy of the driving vehicle can be maintained.

<Modification 1>
FIG. 10 shows a hardware configuration of a position calculation device 100b which is a modification of the position calculation device 100a.
FIG. 11 shows the flow of data of the position calculation device 100b. The position calculation device 100b includes a distance measurement unit 80, such as a radar or a rider, disposed on the side of the autonomous vehicle 200. The distance measuring unit 80 sends a distance measurement signal for distance measurement, such as a radar transmission wave or a laser beam, to the guard rail 8 and measures the distance between the distance measuring unit 80 and the guard rail 8. The marker specifying unit 52a can obtain the distance D in FIG. 2 from the measurement result of the distance measuring unit 80. Specifically, since the mounting position and mounting posture of the distance measuring unit 80 with respect to the autonomous driving vehicle 200 are known, the marker specifying unit 52a can obtain the distance D in FIG. 2 from the measurement result of the distance measuring unit 80. . The marker identifying unit 52a may obtain the distance D shown in FIG. 2 between the sensor axis of the in-vehicle camera 10 of the autonomous driving vehicle 200 and the guard rail 8 from the measured distance.

<Modification 2>
The marker identifying unit 52a also extracts white lines, which are lane lanes present on both sides of the road, from the image captured by the on-vehicle camera 10, and determines the inclination angle of the automated driving vehicle 200 with respect to the extracted lane lanes. You may correct it.

*** Other configuration ***
In the first embodiment, the functions of the image processing unit 51, the associated position calculation unit 52, and the position calculation unit 53 of the position calculation devices 100a and 100b are realized by software, but as another configuration, the position calculation devices 100a and 100b are provided. The function of may be realized by hardware. Although the position calculation device 100b will be described as an example, the position calculation device 100a is also similar to the position calculation device 100a.
FIG. 12 is a diagram showing another configuration of the position calculation device 100b. In the position calculation device 100 b of FIG. 12, the functions of the image processing unit 51, the related position calculation unit 52 and the position calculation unit 53 are realized by the electronic circuit 903. The other configuration of FIG. 12 is the same as the configuration of the position calculation device 100b.

  Specifically, the electronic circuit 903 is a single circuit, a complex circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation of Gate Array. ASIC is an abbreviation of Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array. The functions of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 of the position calculation device 100b may be realized by one electronic circuit, or may be realized by being distributed to a plurality of electronic circuits. As another modification, some functions of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 of the position calculation device 100b may be realized by an electronic circuit, and the remaining functions may be realized by software.

  Each of the processor and electronics is also referred to as processing circuitry. That is, the functions of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 are realized by processing circuitry.

  In the position calculation device 100b, "part" of the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 may be read as a "process". The operations by the image processing unit 51, the related position calculation unit 52, and the position calculation unit 53 can be understood as a position calculation method realized by a position calculation program.

  As described above, the position calculation device according to the first embodiment is a position calculation device mounted on an automatic driving vehicle, and the coordinates of the arranged position are used for automatic driving of the automatic driving vehicle. A camera device for capturing a coordinate marker registered in the high-accuracy map data, and identification information for identifying the coordinate marker based on the image of the coordinate marker captured by the camera device, and based on the identification information The coordinate position of the coordinate marker is obtained from the three-dimensional high-accuracy map data, and the related position calculation unit is used to calculate a marker related position indicating the position of the automatic driving vehicle using the coordinate of the coordinate marker. It features.

  Further, the related position calculation unit may determine an azimuth angle of the coordinate marker with respect to a reference direction from an image of the coordinate marker, and calculate the marker related position using the determined azimuth angle.

  The vehicle further includes a position calculation unit that calculates the position of the autonomous driving vehicle using a positioning signal transmitted by a positioning satellite, and the related position calculation unit calculates the autonomous driving vehicle calculated by the position calculation unit using the positioning signal. A position correction unit may be provided that corrects the position of the image by using the marker related position.

  Furthermore, an inertial measurement device is provided, the position calculation unit calculates the position of the autonomous vehicle from the observation information obtained by the inertial measurement device, and the position correction unit is configured to calculate the position calculation unit from the observation information. The calculated position of the autonomous vehicle may be corrected using the marker related position.

  Thus, according to the first embodiment, since the position of the automatic driving vehicle 200 can be obtained from the coordinate marker 90, the positioning signal from the GNSS satellite can not be received, and even in the area where the roadside system is not installed. , Can maintain the position accuracy of self-driving cars.

  P (m) Marker related position, 3 2D barcode, 4 numerical value, 5 XYZ coordinate system, 6 vector, 7 lane link, 8 guardrail, 10 onboard camera, 20 positioning receiver, 30 IMU, 40 I / O interface, 50 Processor, 51 image processing unit, 52 related position calculation unit, 52a marker specifying unit, 52b position correction unit, 53 position calculation unit, 60 memory, 70 auxiliary storage device, 71 DMD, 72 correspondence information, 80 distance measurement unit, 81 signal Line, 90, 90a, 90b, 90c, 90d Coordinate marker, 91 identification information, 100a, 100b Position calculation device, 200 self-driving car, 901 map data, 902 camera device, 903 electronic circuit.

Claims (7)

A position calculation device mounted on an autonomous vehicle,
A camera device for photographing a coordinate marker registered in map data used for automatic driving of the autonomous driving vehicle, and a coordinate of a position where the camera is arranged;
The identification information for identifying the coordinate marker is extracted based on the image of the coordinate marker captured by the camera device, and the coordinate of the coordinate marker is acquired from the map data based on the identification information, and the coordinates of the coordinate marker And a related position calculation unit that calculates a marker related position indicating the position of the autonomous vehicle using. The related position calculation unit
The position calculation device according to claim 1, wherein an azimuth angle of the coordinate marker with respect to a reference direction is determined from an image of the coordinate marker, and the marker related position is calculated using the determined azimuth angle. The position calculation device further includes
A position calculation unit that calculates the position of the autonomous vehicle using a positioning signal transmitted by a positioning satellite;
The related position calculation unit
The position calculation device according to claim 1, further comprising a position correction unit that corrects the position of the autonomous vehicle calculated by the position calculation unit using the positioning signal using the marker related position. The position calculation device further includes
Equipped with an inertial measurement device,
The position calculation unit
Calculating the position of the autonomous vehicle from the observation information obtained by the inertial measurement device;
The position correction unit
The position calculation apparatus according to claim 3, wherein the position calculation unit corrects the position of the autonomous vehicle calculated from the observation information using the marker related position.   The position calculation device according to any one of claims 1 to 4, wherein the map data is a three-dimensional map. On the computer
Identification that identifies the coordinate marker based on an image of the coordinate marker captured by a camera device that captures a coordinate marker registered in map data used for automatic driving of an automatic driving vehicle, wherein the coordinates of the arranged position are registered in the map data used for automatic driving Processing to extract information;
A process of acquiring coordinates of the coordinate marker from the map data based on the identification information;
A position calculation program for executing a process of calculating a marker related position indicating the position of the automatic driving vehicle using the coordinates of the coordinate marker. An autonomous vehicle equipped with a camera device is disposed in an area where the camera device can capture an image when passing on a road,
The coordinates of the arranged position are registered in the map data used for automatic driving of the self-driving car,
A coordinate marker used in a calculation method for calculating the position of the autonomous vehicle from an azimuth angle with respect to a reference direction when it is photographed by the camera device.

Images