Determination device, determination method, and determination program

Abstract

The present invention provides a determination device, etc., capable of appropriately processing information related to convex mirrors. [Solution] The system includes a LiDAR sensor 205 for detecting objects, and a control unit 201 that identifies the detection direction of the detected object, acquires map information including location information indicating the location of a mirror feature which is the location of a mirror feature, and location information indicating the position of the LiDAR sensor 205, extracts a detection result from the detection results of the LiDAR sensor 205 that is calculated based on the self position and the location of the feature based on the identified detection direction and corresponds to the mirror feature prediction range which includes the location of the feature, and determines that the object included in the extracted detection result is a mirror feature. At this time, the mirror feature prediction range is a preset range that is in the detection direction as seen from the LiDAR sensor 205 and includes the location of the feature.

Description

Technical Field

[0001] This application relates to the technical field of information processing for curve mirrors.

Background Art

[0002] In an autonomous vehicle, it is necessary to match the position of a ground object measured by a sensor such as a LIDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging) sensor with the position of the ground object described in the map data for autonomous driving, and estimate the position of the host vehicle with high accuracy. Ground objects used include signs and billboards, etc. Map data for autonomous driving including the positions of these ground objects needs to be maintained and updated in line with reality in order to perform stable autonomous driving.

[0003] As one technology for this, the development of laser measurement technology that emits laser light onto the ground surface and utilizes the point cloud data obtained by receiving the reflected light for the use and update of map data, etc., is underway. For example, in Patent Document 1, although the point cloud data contains a lot of data other than the road surface such as buildings and street trees, a technology for extracting data that measures the road surface from this is disclosed. By using the technology of Patent Document 1, the road area can be determined from the point cloud data measured on the ground surface. Also, by using the above laser measurement technology including the technology of Patent Document 1, it is also possible to detect ground objects such as signs installed beside the road, etc., other than the above road surface.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Here, on roads, there are sometimes convex mirrors installed to allow drivers to see people or cyclists entering the road from side streets, for example, people or cyclists who are not visible to the driver. Convex mirrors have a high reflectivity of the laser light used in the laser measurement technology described above. Therefore, if the LiDAR sensor emits laser light towards the convex mirror at the same time that a person or other object is reflected in the mirror, the image of the person or object contained in the reflected light may cause the system to detect that the person or object is actually located at the position of the convex mirror. This problem can lead to problems such as unnecessary braking of the autonomous vehicle due to the false detection of the person or object. These problems can be solved by knowing the position of the convex mirror in advance.

[0006] Therefore, this application was made in view of these problems and requirements, and one example of the problem is to provide a determination device and determination method that can appropriately process information using convex mirrors, as well as a program for said determination device. [Means for solving the problem]

[0007] The invention described in claim 1 is a determination device comprising: detection means for detecting an object; identification means for identifying the detection direction of the detected object as seen from the detection means; map information including location information indicating the location of a feature which is the location of a mirror feature having a mirror; and location information indicating the position of the detection means; and determination means for extracting a detection result from the detection results by the detection means that is calculated based on the identified detection direction, the position of the detection means, and the mirror feature prediction range which is a mirror feature prediction range which is a predetermined range that is in the detection direction as seen from the detection means and includes the feature location.

[0008] The invention described in claim 4 comprises: detection means for detecting an object; identification means for identifying the detection direction of the detected object as seen from the detection means; determination means for determining whether the object is a mirror feature having a mirror based on the detected object and the identified detection direction; acquisition means for acquiring map information including location information indicating the location of the mirror feature, and location information indicating the location of the detection means; and transmission means for transmitting to the outside that the mirror feature has disappeared if it is determined that the object at the location indicated by the location information is not the mirror feature based on the acquired map information and location information.

[0009] The invention described in claim 7 comprises: detection means for detecting an object; identification means for identifying the detection direction of the detected object as seen from the detection means; and determination means for determining whether the object is a mirror feature having a mirror, wherein the determination means is a determination device that determines that the object in the identified detection direction is a mirror feature when the detection result from the detection means includes a detection result indicating the mirror of the mirror feature, and when it is determined to be a mirror feature, the determination device ignores the sensor data of reflected light from the mirror of the mirror feature.

[0010] The invention described in claim 11 is a determination method to be performed in a determination device comprising a detection means, a identification means, an acquisition means, and a determination means, the method comprising: a detection step of detecting an object with the detection means; a identification step of identifying the detection direction of the detected object as seen from the detection means with the identification means; an acquisition step of acquiring map information including location information indicating the location of a mirror feature which is the location of a mirror feature having a mirror, and location information indicating the position of the detection means which is the position of the detection means with the acquisition means; and a determination step of extracting a detection result from the detection result by the detection means that is calculated based on the identified detection direction, the position of the detection means and the position of the detection means and corresponds to a mirror feature prediction range which includes the location of the feature, and determining that the object included in the extracted detection result is the mirror feature with the determination means, wherein the mirror feature prediction range is a preset range which is in the detection direction as seen from the detection means and includes the location of the feature.

[0011] The invention described in claim 12 is a determination method performed in a determination device comprising detection means, identification means, determination means, acquisition means, and transmission means, the determination method comprising: detection step of detecting an object with the detection means; identification step of identifying the detection direction of the detected object as seen from the detection means with the identification means; determination step of determining whether the object is a mirror feature having a mirror based on the detected object and the identified detection direction with the determination means; acquisition step of acquiring map information including location information indicating the location of the mirror feature and location information indicating the location of the detection means with the acquisition means; and transmission step of transmitting to the outside with the transmission means that the mirror feature has disappeared if it is determined that the object at the location indicated by the location information is not the mirror feature based on the acquired map information and location information.

[0012] The invention described in claim 13 is a determination method performed in a determination device comprising detection means, identification means, and determination means, the method comprising: a detection step of detecting an object with the detection means; an identification step of identifying the detection direction of the detected object as seen from the detection means with the identification means; and a determination step of determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction, the determination method wherein, in the determination step, if the detection result by the detection means includes a detection result indicating the mirror of the mirror feature, the method determines that the object in the identified detection direction is the mirror feature, and if it is determined to be the mirror feature, the method ignores the sensor data of reflected light from the mirror of the mirror feature.

[0013] The invention described in claim 14 is a determination program that causes a computer to function as a detection means for detecting an object, a identification means for identifying the detection direction of the detected object as seen from the detection means, a map information including location information indicating the location of a feature which is the location of a mirror feature having a mirror, and location information indicating the position of the detection means, and a determination means for extracting a detection result from the detection results by the detection means that is calculated based on the position of the detection means and the position of the feature and includes the location of the feature, and determines that the object included in the extracted detection result is the mirror feature, wherein the mirror feature prediction range is a preset range that is in the detection direction as seen from the detection means and includes the location of the feature.

[0014] The invention described in claim 15 is a determination program that causes a computer to function as: detection means for detecting an object; identification means for identifying the detection direction of the detected object as seen from the detection means; determination means for determining whether the object is a mirror feature having a mirror based on the detected object and the identified detection direction; acquisition means for acquiring map information including location information indicating the location of the feature which is the location of the mirror feature, and location information indicating the location of the detection means; and transmission means for transmitting to the outside that the mirror feature has disappeared if it is determined that the object at the location indicated by the location information is not the mirror feature based on the acquired map information and location information.

[0015] The invention described in claim 16 is a determination program that causes a computer to function as a detection means for detecting an object, a identification means for identifying the detection direction of the detected object as seen from the detection means, and a determination means for determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction, wherein the computer, which functions as the determination means, determines that the object in the identified detection direction is a mirror feature when the detection result from the computer, which functions as the detection means, includes a detection result indicating the mirror of the mirror feature, and the computer, when determined to be a mirror feature, causes the computer to ignore the sensor data of reflected light from the mirror of the mirror feature. [Brief explanation of the drawing]

[0016] [Figure 1] This is a block diagram showing the schematic configuration of the determination device according to the embodiment. [Figure 2] This block diagram shows the outline configuration of the map data management system in the embodiment. [Figure 3] (A) is a diagram illustrating the format of the transmitted data in the embodiment. (B) is a diagram illustrating the format of the map data in the embodiment. [Figure 4]This is a diagram for explaining a method of calculating the prediction range of a curve mirror in an embodiment. [Figure 5] (A) is a flowchart showing an operation example of transmission processing of transmission data by the map data management system of the embodiment. (B) is a diagram showing an example of an actual mirror. [Figure 6] This is a flowchart showing an operation example of processing using information such as a curve mirror in an embodiment. [Figure 7] This is a flowchart showing details of an operation example of processing using information such as a curve mirror in an embodiment.

Mode for Carrying Out the Invention

[0017] A mode for carrying out the invention of the present application will be described with reference to FIG. 1. Note that FIG. 1 is a block diagram showing the schematic configuration of the determination device of this embodiment.

[0018] As shown in FIG. 1, the determination device 200 of this embodiment includes a detection unit 205, a specification unit 201A, an acquisition unit 201B, and a determination unit 201C.

[0019] At this time, the detection unit 205 detects an object. Then, the specification unit 201A specifies the detection direction of the object detected by the detection unit 205 as seen from the detection unit 205. On the other hand, the acquisition unit 201B acquires map information including position information indicating a ground object position, which is the position of a ground object having a mirror, and position information indicating its own position, which is the position of the detection unit 205. Then, the determination unit 201C extracts, from the detection result by the detection unit 205, a detection result corresponding to the mirror ground object prediction range calculated based on the above-described own position and the above-described ground object position and including the ground object position based on the detection direction specified by the specification unit 201A, and determines that the object included in the extracted detection result is the mirror ground object. At this time, the mirror ground object prediction range is a preset range that is in the above-described detection direction as seen from the detection unit 205 and includes the ground object position.

[0020] As described above, according to the operation of the determination device 200 of the present embodiment, based on the detection direction of the object specified by the specifying means 201A, the detection result corresponding to the mirror ground object prediction range that is calculated based on the self-position of the detection means 205 and the ground object position of the mirror ground object and includes the ground object position is extracted from the detection results by the detection means 205, and the object included in the extracted detection result is determined to be a mirror ground object. At this time, the mirror ground object prediction range is a preset range that is in the detection direction as seen from the detection means 205 and includes the ground object position. Therefore, it is possible to appropriately perform processing using information related to the mirror ground object.

Example

[0021] Examples will be described with reference to FIGS. 2 to 7. The examples described below are examples when the present embodiment is applied to the map data management system S. FIG. 2 is a block diagram showing the schematic configuration of the map data management system of the example, FIG. 3 is a diagram exemplifying the format of the map data of the example, etc., and FIG. 4 is a diagram for explaining the calculation method of the curve mirror prediction range of the example. Further, FIG. 5 is a flowchart etc. showing an operation example of the transmission process of transmission data by the map data management system of the example, FIG. 6 is a flowchart showing an operation example of processing using information such as a curve mirror of the example, and FIG. 7 is a flowchart showing the details of the operation example of processing using information such as a curve mirror of the example.

[0022] [1. Configuration and Overview of Map Data Management System S] As shown in FIG. 2, the map data management system S of the present example includes a server device 100 that manages map data and in - vehicle terminals 200 (an example of the determination device 200 of the embodiment) mounted on each of a plurality of vehicles. The server device 100 and each in - vehicle terminal 200 are connected so as to be able to exchange various data via a network NW such as the Internet. In FIG. 2, one in - vehicle terminal 200 is shown, but the map data management system S may include a plurality of in - vehicle terminals 200. Also, the server device 100 may be composed of a plurality of devices.

[0023] In a vehicle equipped with a LiDAR sensor (described later) along with the in-vehicle terminal 200, the LiDAR sensor detects geographical features by receiving reflected light from these features in the vicinity of the vehicle, and transmits data corresponding to the detection result to the server device 100 as transmission data having the data structure of the embodiment described later. In this embodiment, features include so-called curve mirrors installed by the national or local government at necessary locations along the roadside, as well as mirrors installed by an individual, for example, at the boundary between their own parking lot and the road. In the following description, these will be collectively referred to as "curve mirrors, etc." In addition, the vehicle terminal 200 performs the processing described later corresponding to the curve mirrors, etc., based on the reflected light of the laser beam from the curve mirrors, etc.

[0024] The server device 100 updates the map data recorded in the server device 100, as described below, based on the multiple transmission data received from each of the multiple in-vehicle terminals 200, and transmits the map data corresponding to the request to the requesting in-vehicle terminal 200 in response to a request from any of the in-vehicle terminals 100. Note that the updating of the map data may be performed by another device instructed by the server device 100.

[0025] [2. Configuration of the in-vehicle terminal 200] Next, the configuration of the in-vehicle terminal 200 in this embodiment will be described.

[0026] As shown in Figure 2, the in-vehicle terminal 200 is broadly composed of a control unit 201 (an example of the identification means 201A, acquisition means 201B, and determination means 201C in the embodiment), a storage unit 202, a communication unit 203, and an interface unit 204 (an example of the "transmission means" in this application), and is connected to external devices, a LiDAR sensor 205 (an example of the detection means 205 in the embodiment) and an internal sensor 206.

[0027] The storage unit 202 is composed of, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores the OS (Operating System), the processing program of the embodiment, map data, and various other data. The map data contains map location information indicating the position of a feature that is the target of detection by the LiDAR sensor 205, and a feature ID for identifying that feature from other features. In this case, since the map location information and the feature ID are information linked to a single feature, the feature ID can be said to be one of the location information indicating the position of that single feature. Examples of the above-mentioned features include, in addition to the curve mirror in this embodiment, people, other vehicles, traffic lights, signs, buildings, and roadside vegetation that exist around the vehicle on which the in-vehicle terminal 200 is mounted. Furthermore, map data similar to that stored in the memory unit 202 (i.e., map data describing the map location information and feature ID for each feature) is also stored in the memory unit 102 of the server device 100, allowing the same feature to be identified by its feature ID in both the in-vehicle device 200 and the server device 100. Moreover, the map data stored in the memory unit 202 may, for example, store map data for the entire country, or it may store map data corresponding to a certain area including the vehicle's current location, which is received in advance from the server device 100 or the like.

[0028] The communication unit 203 controls the communication status between the in-vehicle terminal 200 and the server device 100.

[0029] The interface unit 204 provides an interface function for exchanging data between the LiDAR sensor 205 and the under-situ sensor 206 and the in-vehicle terminal 200.

[0030] The LiDAR sensor 205 is mounted, for example, on the roof of a vehicle, and as one function, it emits pulsed laser light (more specifically, pulsed infrared laser light) to scan around the vehicle, drawing a circle with the laser light emission point. At this time, the LiDAR sensor 205 emits laser light downwards from the roof at a certain angle. It then receives the light reflected from points on the surfaces of various objects around the vehicle and generates reflection intensity data showing the reflection intensity at each point, as well as point cloud distance data. This reflection intensity data shows the reflection intensity of the laser light from the ground and objects. The point cloud distance data shows the direction in which the laser light was emitted and the distance to the point of laser light emission on the ground or objects in that direction. This reflection intensity data and point cloud distance data are collectively called sensor data. In addition, multiple LiDAR sensors 205 may be mounted on the front or rear of the vehicle, and the reflection intensity data and point cloud distance data acquired by each sensor may be combined to generate sensor data for the area around the vehicle.

[0031] The LiDAR sensor 205 immediately generates sensor data upon measuring the reflection intensity and transmits it to the in-vehicle terminal 200 via the interface unit 204. When the control unit 201 receives the sensor data from the LiDAR sensor 205, it stores the received sensor data in the storage unit 202, associating it with measurement position information indicating the vehicle's position at the time of sensor data reception (i.e., the position of the in-vehicle terminal 200 including the LiDAR sensor 205) and measurement date and time information indicating the date and time of sensor data reception. The control unit 201 may delete from the storage unit 202 any sensor data, measurement position information, and measurement date and time information that have been stored in the storage unit 202 for a predetermined time after measurement, or that have been transmitted to the server device 100.

[0032] The internal sensors 206 are a general term for sensors mounted on a vehicle, including satellite positioning sensors (GNSS (Global Navigation Satellite System)), gyro sensors, and vehicle speed sensors.

[0033] The control unit 201 consists of a CPU (Central Processing Unit) that controls the entire control unit 201, a ROM (Read Only Memory) in which control programs for controlling the control unit 201 are pre-stored, and a RAM (Random Access Memory) for temporarily storing various data. The CPU then reads and executes various programs stored in the ROM and the memory unit 202 to realize various functions, including the transmission processing of data transmitted by the map data management system S of the embodiment and processing using information such as curve mirrors of the embodiment.

[0034] The control unit 201 acquires estimated vehicle position information. The estimated vehicle position information may be generated by a device outside the in-vehicle terminal 200, or it may be generated by the control unit 201. The estimated vehicle position information can be generated, for example, by matching the location of features measured by the LiDAR sensor 205 with the location of features in the map data for autonomous driving, or by generating it based on information detected by the interior sensor 206 and the map data, or by a combination of these methods.

[0035] Furthermore, the control unit 201 detects, for example, a curve mirror or other feature present on the side of the road based on the estimated vehicle position information and sensor data from the LiDAR sensor 205, and transmits transmission data 1 corresponding to the detection result to the server device 100 using the data structure of transmission data 1 in the embodiment. This curve mirror detection process will be described in detail later with reference to Figure 5.

[0036] Here, the data structure of the transmitted data 1 in the embodiment will be explained using Figure 3(A).

[0037] The transmission data 1 for each object in the embodiment, including the curve mirror, has a data structure consisting of basic information TMB, recognized object information TMO, and specific information TMS, as illustrated in Figure 3(A).

[0038] The basic information TMB consists of a header TMB1, vehicle metadata TMB2, and location data TMB3. Header TMB1 is data indicating the version of the data format for transmission data 1 and a timestamp indicating when the data for transmission data 1 was transmitted. Vehicle metadata TMB2 includes a vehicle ID for identifying the vehicle on which the in-vehicle device 200 is installed from other vehicles, data indicating the size of the vehicle, and data indicating the type of sensor connected to the in-vehicle terminal 200. It is assumed that, in addition to the LiDAR sensor 205 in this embodiment, the in-vehicle terminal 200 also has a camera that captures images of the surroundings and outputs image data. Location data TMB3 is data indicating the position of the vehicle on which the in-vehicle terminal 200 is installed at the time the in-vehicle terminal 200 detects and recognizes an object (hereinafter, the object will be appropriately referred to as "object"). This position is generally indicated by latitude and longitude, and may also include altitude data.

[0039] Next, the recognized object information TMO consists of object ID data TMO1, object type data TMO2, detection date and time data TMO3, location data TMO4, shape data TMO5, and size data TMO6. In this case, object ID data TMO1 is ID data for identifying the feature whose specifications etc. are transmitted by transmission data 1 (hereinafter, the feature will be appropriately referred to as the "transmission target feature") from other features. If the detected feature is the first time it has been detected, this object ID data TMO1 may be blank data. Object type data TMO2 is code data for identifying the type of transmission target feature (for example, types such as "traffic sign," "traffic light," "curve mirror," etc., and in the case of "curve mirror," the type may be divided according to the number of mirrors attached to the pole), and in this example, a value indicating that it is a curve mirror, etc. is set. Detection date and time data TMO3 is data indicating the date and time when the transmission target feature was detected using the LiDAR sensor 205. Location data TMO4 is data indicating the latitude, longitude, and height (elevation) detected as the location of the transmitted object. Shape data TMO5 is data indicating the shape of the transmitted object (e.g., round shape, square shape, etc.). Finally, size data TMO6 is data indicating the length and width of the feature part of the transmitted object. In the case of a convex mirror, the feature part is the mirror itself. Furthermore, the width in size data TMO6 is described using the positions of both ends of a horizontal line segment that is parallel to the mirror surface and has the same length as the diameter of the mirror.

[0040] Furthermore, the specific information TMS is added as information specific to convex mirrors, etc., and consists of position and height data TMS1, curvature data TMS2, and reflection presence / absence data TMS3. In this case, the position and height data TMS1 is data indicating the height of the mirror portion of the convex mirror, etc., from the ground surface. The curvature data TMS2 is data indicating the curvature of the mirror. The reflection presence / absence data TMS3 is data indicating whether or not the convex mirror, etc., functioned as a mirror.

[0041] Meanwhile, the control unit 201 predicts the actual position of the convex mirror, etc., as seen from the vehicle (i.e., the LiDAR sensor 205), based on the estimated vehicle position information and the map position information of the convex mirror, etc., shown in the map data. At this time, the control unit 201 calculates and sets a prediction range that includes the position of the convex mirror, etc., with a certain margin. In the following description, the prediction range that includes the position of the convex mirror, etc., in the embodiment will be simply referred to as the "convex mirror prediction range".

[0042] Here, we will specifically explain how to set the curve mirror prediction range using Figure 4. The coordinate system in Figure 4 is as follows. Note that Figure 4 shows the curve mirror prediction range which includes the positions of the curve mirrors included in the curve mirrors of the embodiment.

[0043] Map coordinate system: Xm, Ym Vehicle coordinate system: XV, YV Map coordinate system for the location of the curve mirror: mxm,mym Predicted position of curve mirror in vehicle coordinate system: lxv, lyv Estimated vehicle position in map coordinate system: xm, ym Estimated vehicle azimuth angle in map coordinate system: Ψm The above-mentioned vehicle coordinate system refers to a coordinate system that uses the position of the vehicle on which the vehicle body terminal 200 is mounted as the reference point (origin).

[0044] The control unit 201 calculates the corresponding curve mirror prediction range from the curve mirror map position information of a curve mirror located in the direction of travel of the vehicle (for example, 10m ahead), based on the estimated vehicle position information. Here, as shown in Figure 4, we will specifically explain the case where a curve mirror located on the left side of the road (lane) on which the vehicle V is traveling is detected, and the curve mirror prediction range including the position of the said curve mirror is calculated. First, the control unit 201 calculates the curve mirror prediction position 301 based on the curve mirror map position and the estimated vehicle position. The curve mirror prediction position is obtained by the following equation (1).

[0045]

number

[0046] The control unit 201 stores the extracted sensor data 321 in the storage unit 202, associating it with the feature ID corresponding to the location of the curve mirror on the map and the measurement date and time information corresponding to the sensor data from which the sensor data 321 was extracted. The processing of an embodiment using information such as a curve mirror, including the sensor data 321 obtained in this way, will be described in detail later with reference to Figure 6.

[0047] [3. Configuration of Server Device 100] Next, the configuration of the server device 100 will be described. As shown in Figure 2, the server device 100 is broadly composed of a control unit 101, a storage unit 102, a communication unit 103, a display unit 104, and an operation unit 105.

[0048] The storage unit 102 is composed of, for example, an HDD or SSD, and stores the OS, the map data, the transmission data received from the in-vehicle terminal 200, and various other data.

[0049] The communication unit 103 controls the communication status with the in-vehicle terminal 200.

[0050] The display unit 104 is composed of, for example, a liquid crystal display, and displays information such as characters and images.

[0051] The control unit 105 is composed of, for example, a keyboard, a mouse, etc., and receives operation instructions from the operator and outputs the content of those instructions as instruction signals to the control unit 101.

[0052] The control unit 101 consists of a CPU that controls the entire control unit 101, a ROM in which control programs for controlling the control unit 101 are pre-stored, and a RAM for temporarily storing various data. The CPU then reads and executes various programs stored in the ROM and the storage unit 102 to realize various functions, including the processing of receiving transmitted data by the map data management system S of this embodiment.

[0053] The control unit 101 receives multiple transmission data received from one or more in-vehicle terminals 200, and uses them to update the map data corresponding to the corresponding features.

[0054] Here, the data structure of the map data in the example will be explained using Figure 3(B).

[0055] The map data FM stored in the in-vehicle terminal 200 and server device 100 of the embodiment has a data structure consisting of basic information FMB and specific information FMS, as illustrated in Figure 3(B).

[0056] The basic information FMB consists of object ID data FMB1, object type data FMB2, location data FMB3, and size data FMB4. In this case, object ID data FMB1 is ID data used to distinguish the corresponding feature from other features. Object type data FMB2 is code data used to identify the type of the corresponding feature, similar to the object type data TMO2 mentioned above, and in this example, a value indicating "curve mirror, etc." is set. Location data FMB3 is data showing the latitude, longitude, and height (elevation) of the location where the corresponding feature exists. Size data FMB4 is data showing the vertical and horizontal dimensions of the characteristic part of the corresponding feature, similar to the size data TMO6 mentioned above. In addition to location data FMB3, information indicating which road link the feature belongs to may be added.

[0057] Next, the unique information FMS is information added as information specific to the curve mirror, etc., and consists of curvature data FMS1, observation target data FMS2, shape data FMS3, owner data FMS4, and unusable time period data FMS5. In this case, curvature data FMS1 is data that indicates the curvature of the mirror when the corresponding feature is a curve mirror, etc., similar to the curvature data TMS2 above. Observation target data FMS2 is data that indicates the object reflected in the mirror (for example, a merging side road, etc.) when the corresponding feature is a curve mirror, etc., and specifically it is road ID data for identifying the reflected road, etc. from other roads, etc. It is preferable to register the object reflected in the mirror in association with the direction in which the curve mirror, etc. is viewed (or road ID data indicating the road when the curve mirror, etc. is viewed). Shape data FMS3 is data that indicates the shape of the corresponding feature, similar to the shape data TMO5 above. Ownership data FMS4 indicates the owner of the corresponding feature (for example, whether it is owned by the national or local government, or by an individual) when the feature is a convex mirror, etc. Unavailable time period data FMS5 indicates the time period when, for example, the sun is behind the convex mirror, etc. as viewed from a vehicle, the LiDAR sensor 205 cannot detect objects on the road, etc., as indicated by the observation target data FMS2, using the convex mirror, etc. Note that unavailable time period data FMS5 may be set for each type of sensor that may be mounted on the vehicle. For example, for a camera as a sensor, the time period when morning dew adheres to the mirror is assumed to be unavailable time period data FMS5. Also, for example, if part of the convex mirror, etc. is hidden by street trees depending on the season, data indicating that season may be set as unavailable time period data FMS5. Furthermore, among the above-mentioned specific information FMS, the observation target data FMS2, owner data FMS4, and unusable time period data FMS5 can be configured to be described in the map data FM based on information from the installer when the corresponding convex mirror, etc., is installed.

[0058] [4. Example of operation of the map data management system S] [4.1. Examples of operation when detecting curve mirrors, etc.] Next, using the flowchart in Figure 5(A), we will explain an example of the operation of the map data management system S, specifically the operation when an in-vehicle terminal 200 detects a curve mirror or the like. In the flowchart in Figure 5, we explain the process in which one in-vehicle terminal 200 generates the above transmission data 1 using the detection result from its LiDAR sensor 205 and transmits it to the server device 100. However, the same process is performed for each in-vehicle terminal 200 included in the map data management system S. Furthermore, the processes of the in-vehicle terminal 200 in Figure 5(A) from step S101 to step S105 are executed periodically (for example, every predetermined time and / or every time the vehicle on which the in-vehicle terminal 200 is installed travels a predetermined distance). Upon receiving the processing of step S105 from the in-vehicle terminal 200, the server device 100 executes the processes of steps S201 to S202.

[0059] First, the control unit 201 of the in-vehicle terminal 200 acquires estimated vehicle position information (step S101).

[0060] Next, the control unit 201 recognizes a feature that appears to be a convex mirror or the like (see Figure 4) located along the direction of movement of the vehicle on which the in-vehicle terminal 200 is mounted, based on sensor data from the LiDAR sensor 205 (step S102). Specifically regarding step S102, in general convex mirrors, as illustrated in Figure 5(B), the pole PL to which the mirrors MR1 ​​and MR2 are fixed is often equipped with a mark MK such as "Caution" with an arrow. By detecting the mark MK based on the sensor data, the control unit 201 can recognize that the feature to which the mark MK is attached is a convex mirror or the like. Also, for example, the control unit 201 can recognize that a feature is a convex mirror or the like based on the attributes of an object detected at a position that cannot be detected depending on the emission angle of the laser beam from the LiDAR sensor 205. More specifically, for example, if a vehicle is detected by laser beam emitted upward from the horizontal direction, the control unit 201 can determine that the vehicle is a vehicle reflected in a mirror such as a convex mirror, and thereby the control unit 201 can recognize that the position of the vehicle is the position of the mirror such as a convex mirror. Next, the control unit 201 obtains the map location information of the feature (a feature that appears to be a curve mirror, etc.) recognized in step S102 from the map data FM corresponding to the estimated vehicle position indicated by the estimated vehicle position information obtained in step S101 (step S103). At this time, as described above, the control unit 201 obtains the map location information and feature ID of the curve mirror, etc., located in the direction of travel of the vehicle.

[0061] Next, based on the sensor data from step S102, the control unit 201 determines the curvature, shape, and height from the ground of the curve mirror, etc., recognized in step S102 (step S104). At this time, if, for example, a white line is reflected in the curve mirror, etc., recognized in step S102, the control unit 201 can determine the curvature of the mirror from the degree of distortion. Subsequently, the control unit 201 uses the data obtained up to step S104 to generate transmission data 1 having the data structure exemplified in Figure 3(A) and transmits it to the server device 100. After that, the control unit 201 finishes the generation and transmission process of transmission data 1 as an in-vehicle terminal 200.

[0062] In response, when the control unit 101 of the server device 100 receives the transmission data 1 from the in-vehicle terminal 200 (step S201), it uses it to update the map data FM recorded in the storage unit 102 (step S202). After that, the control unit 101 terminates the map data update process. As a result, the storage unit 102 of the server device 100 stores map data FM updated by multiple transmission data 1 sent from each of the multiple in-vehicle terminals 200.

[0063] If, in step S103, the map location information of a feature cannot be obtained from the map data FM, the feature recognized in step S102 is determined to be a newly installed feature (in this case, a convex mirror, etc.), the feature ID is left blank, and the control unit 101 of the server device 100, which has received the transmission data 1 for the convex mirror, etc., registers the information of the new convex mirror, etc. in the map data FM and updates the map data FM.

[0064] [4.2. Example of operation of processing using information such as curve mirrors] Next, using the flowcharts in Figures 6 and 7, an example of the operation of the processing using information such as a curve mirror in the embodiment by the vehicle terminal 200 will be explained. Furthermore, the processing using information such as a curve mirror in this embodiment is performed periodically (for example, at predetermined intervals and / or each time the vehicle on which the vehicle terminal 200 is installed travels a predetermined distance).

[0065] First, the control unit 201 of the vehicle terminal 200 acquires estimated vehicle position information in the same manner as in step S101 of Figure 5(A) (step S211). Next, the control unit 201 acquires position data FMB3, object ID data FMB1, and other information from the map data FM for curve mirrors and the like that exist near the vehicle, indicated by the estimated vehicle position information acquired in step S211 (step S212; see Figure 3(B)).

[0066] Next, the control unit 201 calculates and sets the curve mirror prediction range from the estimated vehicle position indicated by the estimated vehicle position information, the map position of the curve mirror etc. indicated by the position data FMB3 for the curve mirror etc., and the height of the curve mirror etc. indicated by the position data FMB3 (step S213. See Figure 4).

[0067] Next, the control unit 201 extracts the sensor data detected by the LiDAR sensor 205 that falls within the curve mirror prediction range set in step S213, and determines whether or not laser light was emitted toward the curve mirror, etc., based on the extraction result (step S214). More specifically, the control unit 201 first identifies sensor data detected by emitting laser light within a range that includes the curve mirror prediction range, based on the measurement position of the measurement position information stored in association with the sensor data and the emission angle when the LiDAR sensor 205 emits laser light. Then, the control unit 201 extracts the portion of the identified sensor data that corresponds to the height of the curve mirror, etc. (i.e., sensor data of reflected light from the mirror of the curve mirror, etc.; the same applies hereinafter). For example, the control unit 201 extracts a portion corresponding to the height of the curve mirror, etc., obtained as the difference between the height to the position of the LiDAR sensor 205 on the vehicle obtained in the vehicle coordinate system and the height to the mirror, etc., in the map coordinate system, from the portion corresponding to the azimuth angle θ (see Figure 4) of the curve mirror prediction range based on the direction of the vehicle. Then, based on the extraction result, the control unit 201 determines whether or not there is laser light emitted toward the curve mirror, etc. If, in the determination in step S214, there is no laser light emitted toward the curve mirror, etc. (step S214: NO), the control unit 201 returns to step S211 and repeats the process described above. On the other hand, if, in the determination in step S214, there is laser light emitted toward the curve mirror, etc. (step S214: YES), the control unit 201 then determines whether or not there is a curve mirror, etc., at the point of emission of the laser light, based on sensor data detected from the reflected light of the laser light (step S215). In other words, the control unit 201 can determine, for example, whether a curved mirror or the like (or a mirror of the sort) is present at the laser beam emission point using the same method as performed in step S102 described above.In the determination in step S215, if there is no curve mirror or the like at the laser beam emission point (step S215: YES), the control unit 201 transmits to the server device 100, along with the object ID data FMB1 acquired in step S212, that the curve mirror or the like has disappeared for some reason (step S216). If the determination in step S215 is "YES", for example, the curve mirror or the like is owned by an individual and has been removed by that individual. In other words, if the curve mirror is owned by an individual, the information that the curve mirror has disappeared can be treated as highly reliable. On the other hand, if the curve mirror was installed by the national or local government, it is judged that the possibility of the curve mirror disappearing is low, and it can be determined that it has been removed when a predetermined number of similar reports are collected from other vehicles. As a result, the server device 100 performs processing such as deleting the information of the curve mirror or the like indicated by the transmitted object ID data FMB1 from the map data FM. After that, the control unit 201 terminates the processing using the information of the curve mirror or the like in the embodiment.

[0068] On the other hand, in the determination in step S215, if a convex mirror or the like is present at the laser beam emission point (step S215: NO), the control unit 201 performs processing using sensor data of the reflected light from the convex mirror or the like (step S217). Specifically, the processing in step S217 may include, for example, ignoring the sensor data of the reflected light from the convex mirror or the like in order to avoid processing objects reflected in the mirror or the like. Alternatively, for example, there may be processing to detect vehicles or the like reflected in the mirror of the convex mirror or the like based on the sensor data of the reflected light from the mirror or the like.

[0069] Here, the process of detecting vehicles reflected in a mirror such as a convex mirror, which is the process of step S217, will be specifically explained using Figure 7. That is, as shown in Figure 7, in the process of detecting vehicles reflected in a mirror such as a convex mirror, which is the process of step S217, the control unit 201 determines whether or not a vehicle has been detected as a feature based on the sensor data of the reflected light from the convex mirror, etc. (step S2171). If no vehicle is detected in the determination in step S2171 (step S2171: NO), the control unit 201 returns to Figure 6 and terminates the process using the information of the convex mirror, etc. of the embodiment. On the other hand, if a vehicle is detected in the determination in step S2171 (step S2171: YES), the control unit 201 determines the direction of emission of the laser light from which the reflected light of that reflection intensity was obtained (step S2172). In the determination in step S2172, if the direction of emission of the laser light corresponds to the direction of a mirror such as a convex mirror (step S2172: upward), the control unit 201 determines that the vehicle detected in step S2171 is a vehicle reflected in a mirror such as a convex mirror, and performs a corresponding process, such as deceleration of the vehicle (step S2174). At this time, by referring to the observation target data FMS2 of the convex mirror identified by the object ID data FMB1 acquired in step S212, it is possible to determine which road the vehicle is approaching from. After that, the control unit 201 returns to Figure 6 and terminates the processing using the information of the convex mirror, etc. in the embodiment. On the other hand, in the determination in step S2172, if the direction of emission of the laser light is not in a position corresponding to the direction of the mirror (step S2172: other than upward), the control unit 201 determines that the vehicle detected in step S2171 is a normal vehicle, for example, another vehicle moving on the road on which a vehicle equipped with a vehicle terminal 200 is moving or in the opposite lane, and performs a corresponding process, such as lane change processing (step S2173). After that, the control unit 201 returns to Figure 6 and terminates the processing using information such as the curve mirror in the embodiment.

[0070] As described above, the transmission data 1 in the map data management system S of this embodiment includes position data TMO4 indicating the location of a feature such as a convex mirror, and object type data TMO2 indicating that the feature is such as a convex mirror. The vehicle terminal 200 uses this data to associate the position data TMO4, which indicates the location of the convex mirror detected by the LiDAR sensor 205, with the object type data TMO2, which indicates that the feature identified by the position data TMO4 is such as a convex mirror, and transmits it to the server device 100 for recording.

[0071] Therefore, it is possible to provide a data structure for the transmission data 1 when it is transmitted to the server device 100 to record information about a geographical feature such as a convex mirror.

[0072] Furthermore, the transmitted data 1 is used in the process where, after the vehicle terminal 200 recognizes the position of a feature detected by the LiDAR sensor 205 and recognizes that the detected feature is a convex mirror or the like (see step S102 in Figure 5(A)), it associates the location data TMO4 of the feature with the object type data TMO2 of the feature and transmits it to the server device 100 for recording. Therefore, it is possible to provide a data structure for the transmitted data 1 when it is transmitted to the server device 100 in order to accurately record information about a feature that is a convex mirror or the like.

[0073] Furthermore, when the vehicle terminal 200 obtains object ID data TMO1, which is identification information that allows the server device 100 to identify a feature detected by the LiDAR sensor 205, from the map data FM stored in the storage unit 202, the transmission data 1 is used in the process of associating the location data TMO4, object type data TMO2, and object ID data TMO1 and transmitting them to the server device 100 for recording. Thus, it is possible to provide a data structure for transmission data 1 when transmitting it to the server device 100 in order to accurately record information about the feature in a state that allows for easy identification of the feature.

[0074] Furthermore, since the transmitted data 1 includes shape data TMO5 for curve mirrors, etc., associated with object type data TMO2, it is possible to provide a data structure for transmitted data 1 when transmitting it to the server device 100 in order to record information about features such as curve mirrors in more detail and accurately.

[0075] Furthermore, since the transmitted data 1 includes curvature data TMS2, shape data TMO5, and size data TMO6, etc., it is possible to provide a data structure for the transmitted data 1 when it is transmitted to the server device 100 in order to record information about features such as curve mirrors in more detail and accurately.

[0076] Furthermore, since the transmitted data 1 includes detection date and time data TMO3 for objects such as curve mirrors, associated with object type data TMO2, it is possible to provide a data structure for the transmitted data 1 when it is transmitted to the server device 100 to record information about features such as curve mirrors in correspondence with date and time.

[0077] Furthermore, the control unit 201 identifies the detection direction of the object detected by the LiDAR sensor 205 (the direction of emission of the laser light reflected by the object), and determines whether the object is a convex mirror or the like based on the detected object and the identified detection direction (see Figure 7). As a result, the vehicle terminal 200 can easily recognize the presence of an object such as a vehicle that is in a position where the laser light from the LiDAR sensor 205 cannot be directly shone.

[0078] In the above-described embodiment, the case in which a LiDAR sensor 205 is connected to the vehicle terminal 200 was explained. However, it is also possible to apply this invention to a case in which a camera equipped with, for example, a CMOS (Complementary Metal Oxide Semiconductor) sensor capable of capturing visible light is connected to the vehicle terminal 200, and features are detected using the image data from the camera. In this case, the sensor data handled is an image or video, and recognition object information (TMO) and specific information (TMS), such as the type, location, and size of the features, can be recognized through image recognition processing.

[0079] Furthermore, programs corresponding to the flowcharts shown in Figures 5 to 7 can be recorded on a recording medium such as an optical disc or hard disk, or acquired and recorded via a network such as the Internet, and then read and executed by a general-purpose microcomputer, thereby allowing the microcomputer to function as the control unit 101 or control unit 201 in the embodiment. [Explanation of Symbols]

[0080] 1. Data to be transmitted 1A Location information 1B Mirror Information S Map Data Management System 100 Server Devices 101, 201 Control Unit 201A Specific means 201B Acquisition method 201C Judgment means 200 In-vehicle terminals 205 Detection means (LiDAR sensor) TMB basic information TMO Recognition Object Information TMS specific information TMB1 Header TMB2 Vehicle Metadata TMB3 Location Data TMO1 Object ID Data TMO2 Object Type Data TMO3 detection date and time data TMO4 Location Data TMO5 Shape Data TMO6 size data TMS1 Position and Height Data TMS2 curvature data TMS3 Reflectance Data

Claims

1. A detection means for detecting an object, A means for identifying the detection direction of the detected object as seen from the detection means, An acquisition means that acquires map information including location information indicating the location of a feature which is the location of a mirror feature having a mirror, and location information indicating its own location which is the location of the detection means. A determination means that, based on the identified detection direction, extracts a detection result from the detection results of the detection means that is calculated based on the self position and the object position and corresponds to the mirror object prediction range including the object position, and determines that the object included in the extracted detection result is the mirror object, Equipped with, The determination device is characterized in that the mirror feature prediction range is a preset range that lies in the detection direction as seen from the detection means and includes the location of the feature.

2. In the determination device according to claim 1, A determination device characterized by ignoring sensor data of reflected light from the mirror of a mirror object when it is determined to be a mirror object.

3. In the determination device according to claim 1, A determination device characterized in that, when the detection result by the detection means includes a detection result corresponding to a vehicle, and the object determined to be the mirror object is located in the specified detection direction, the determination means determines that the detection result corresponding to the vehicle is a detection result corresponding to the vehicle reflected in the mirror.

4. A detection means for detecting an object, A means for identifying the detection direction of the detected object as seen from the detection means, A determination means for determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction, An acquisition means that acquires map information including location information indicating the location of the feature, which is the location of the mirror feature, and location information indicating the location of the detection means. Based on the acquired map information and location information, if it is determined that the object at the location indicated by the location information is not the mirror feature, a transmission means transmits to the outside that the mirror feature has disappeared. A determination device characterized by comprising:

5. In the determination device according to claim 4, The determination device is characterized in that the determination means determines that the object in the specified detection direction is the mirror object when the detection result of the detection means includes a detection result indicating an object that is considered to exist in the specified detection direction by being reflected in the mirror.

6. In the determination device according to claim 4 or claim 5, The determination device is characterized in that, when the determination means determines that the object is the mirror feature, it identifies the location of the mirror feature based on the map information and the location information indicating the position of the detection means when the mirror feature was detected.

7. A detection means for detecting an object, A means for identifying the detection direction of the detected object as seen from the detection means, A determination means for determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction, Equipped with, The determination means is a determination device that determines that the object in the specified detection direction is the mirror feature when the detection result from the detection means includes a detection result indicating the mirror of the mirror feature, A determination device characterized by ignoring sensor data of reflected light from the mirror of a mirror object when it is determined to be a mirror object.

8. In the determination device according to claim 7, The determination device is characterized in that the detection result indicating the mirror corresponds to a warning mark indicating that the mirror should be handled with care.

9. In the determination device according to any one of claims 1 to 8, The determination device is characterized in that, when the determination means determines that the object is the mirror feature, it determines the curvature of the mirror, the shape of the mirror, and the height of the mirror, respectively, based on the detection result by the detection means.

10. In the determination device according to claim 9, The determination means is characterized in that, when the detection result by the detection means includes the detection result of a white line on the road reflected in the mirror, the determination means determines the curvature of the mirror based on the detection result of the white line.

11. A determination method to be performed in a determination device comprising detection means, identification means, acquisition means, and determination means, A detection step in which an object is detected by the aforementioned detection means, A process of determining the detection direction of the detected object as seen from the detection means by the identification means, An acquisition step in which the acquisition means acquires map information including location information indicating the location of a feature which is the location of a mirror feature having a mirror, and location information indicating the self-position which is the location of the detection means. A determination step in which, based on the identified detection direction, a detection result is extracted from the detection results by the detection means that is calculated based on the self position and the object position and corresponds to the mirror object prediction range including the object position, and the determination means determines that the object included in the extracted detection result is the mirror object, Includes, The determination method is characterized in that the mirror feature prediction range is a preset range that lies in the detection direction as seen from the detection means and includes the location of the feature.

12. A determination method performed in a determination device comprising detection means, identification means, determination means, acquisition means, and transmission means, A detection step in which an object is detected by the aforementioned detection means, A process of determining the detection direction of the detected object as seen from the detection means by the identification means, A determination step in which the determination means determines whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction. An acquisition step in which the acquisition means acquires map information including location information indicating the location of the feature which is the location of the mirror feature, and location information indicating the location of the detection means. Based on the acquired map information and location information, if it is determined that the object at the location indicated by the location information is not the mirror feature, a transmission step is performed in which the transmission means transmits to the outside that the mirror feature has disappeared. A determination method characterized by including the following.

13. A determination method performed in a determination device comprising detection means, identification means, and determination means, A detection step in which an object is detected by the aforementioned detection means, A process of determining the detection direction of the detected object as seen from the detection means by the identification means, A determination step in which the determination means determines whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction. Includes, In the determination step, if the detection results from the detection means include a detection result indicating the mirror of the mirror feature, the determination method determines that the object in the specified detection direction is the mirror feature. A determination method characterized by ignoring the sensor data of reflected light from the mirror of a mirror object when it is determined to be a mirror object.

14. Computers, detection means for detecting objects, A means for identifying the detection direction of the detected object as seen from the detection means, An acquisition means for acquiring map information including location information indicating the location of a feature which is the location of a mirror feature having a mirror, and location information indicating the position of the detection means, and A determination means extracts a detection result from the detection results of the detection means that is calculated based on the self position and the object position and corresponds to the mirror object prediction range including the object position, and determines that the object included in the extracted detection result is the mirror object. A determination program that functions as such, The determination program is characterized in that the mirror feature prediction range is a preset range that lies in the detection direction as seen from the detection means and includes the location of the feature.

15. Computers, detection means for detecting objects, A means for identifying the detection direction of the detected object as seen from the detection means, A determination means for determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction. An acquisition means for acquiring map information including location information indicating the location of the mirror feature, and location information indicating the location of the detection means, Based on the acquired map information and location information, if it is determined that the object at the location indicated by the location information is not the mirror feature, a transmission means transmits to the outside that the mirror feature has disappeared. A determination program characterized by functioning as such.

16. Computers, detection means for detecting objects, A means for identifying the detection direction of the detected object as seen from the detection means, and A determination means for determining whether the object is a mirror feature having a mirror, based on the detected object and the identified detection direction. A determination program that functions as such, The determination program causes the computer, which functions as the determination means, to determine that the object in the identified detection direction is the mirror feature when the detection results from the computer, which functions as the detection means, include a detection result indicating the mirror of the mirror feature. A determination program characterized in that, when it is determined to be the aforementioned mirror feature, the computer is configured to ignore the sensor data of reflected light from the mirror of the mirror feature.

Images