JP7724673B2 - Identification device and information provision system - Google Patents

Description

The present invention relates to an identification device that distinguishes between target vehicles and other objects at equipment (roadside) installed along a road, and an information provision system equipped with the identification device.

For example, a driving assistance device is known that comprehensively predicts the paths of multiple vehicles traveling in a specific area and provides driving assistance by capturing geometric features through image analysis of images taken by installed cameras, etc., and recognizing vehicles by referencing past data, etc. (see Patent Document 1).

However, using only information acquired by installed cameras, etc., it may be difficult to distinguish between the vehicle's own vehicle and another vehicle, for example, if they are detected in the same or nearly the same position. Without prior information, such as the past data in Patent Document 1, the vehicle's own vehicle may not be identified correctly and may be recognized incorrectly.

Japanese Patent Application Laid-Open No. 2017-207813

The present invention was made in consideration of the above points, and aims to provide an identification device that can reliably distinguish between vehicles and other objects, and an information provision system equipped with the identification device.

To achieve the above objective, the identification device includes an object information acquisition unit that detects objects present in a monitoring area and acquires object location information indicating their location; a vehicle information reception unit that receives self-location information transmitted from the vehicle; a matching unit that time-synchronizes the object location information with the self-location information and compares their positions; and an identification unit that distinguishes the vehicle from other objects in the monitoring area based on the matching results from the matching unit.

The above-mentioned identification device time-synchronizes object position information for the monitored area with self-position information from the vehicle, and then performs position matching between them. In other words, by time-synchronizing information from the vehicle and information from the roadside and then performing position matching, it is possible to reliably identify vehicles and other objects in the monitored area.

In a specific aspect of the present invention, the matching unit compares the spatial coordinates in the time-synchronized object position information and self-position information to identify the vehicle's location within the monitoring area. In this case, time synchronization enables highly accurate comparison of matching or similar spatial coordinates, enabling accurate identification.

In another aspect of the present invention, the object information acquisition unit and the vehicle have a GNSS (Global Navigation Satellite System) receiver, and the matching unit performs time synchronization based on satellite time information included in the object position information and the vehicle's own position information. In this case, time synchronization based on satellite time information is possible.

In yet another aspect of the present invention, the vehicle is an autonomous vehicle, and the vehicle information receiving unit receives future position information of the autonomous vehicle as its own position information. In this case, various decisions and information can be made and provided from the roadside based on the future position information.

In yet another aspect of the present invention, the object information acquisition unit includes a sensor unit that senses the monitoring area and acquires object position information, and extracts moving objects present in the monitoring area based on the object position information from the sensor unit, and the identification unit identifies vehicles from among the moving objects extracted by the object information acquisition unit. In this case, the vehicle to be extracted can be identified based on sensing by the sensor unit.

In yet another aspect of the present invention, a determination unit is provided that determines whether a collision has occurred between the vehicle and a moving object other than the vehicle, based on the results of the identification by the identification unit. In this case, it becomes possible to provide driving assistance from the roadside to avoid a collision that has been detected.

To achieve the above objective, an information provision system includes any of the above-mentioned identification devices, and transmits analysis results regarding object position information, including identification by the identification unit, to a vehicle.

By including an identification device, the information provision system described above time-synchronizes object position information for the monitored area with self-position information from the vehicle, and then performs position matching between them. In other words, by time-synchronizing information from the vehicle and information from the roadside and then performing position matching, it is possible to reliably identify vehicles and other objects in the monitored area, and transmit information about the identification results to the vehicle.

1 is a conceptual diagram for explaining an overview of an identification device according to a first embodiment and an information providing system including the same; 1 is a conceptual diagram showing how a detection is performed in a monitoring area in an identification device and an information providing system. 1 is a block diagram showing an example of the overall configuration of an information providing system; FIG. 2 is a block diagram showing a detailed configuration example of an information providing system. 10A and 10B are data diagrams showing an example of an outline of target information as a detection result in a monitoring area. 10A and 10B are data diagrams showing an example of an outline of communication content. 10A to 10D are conceptual diagrams for explaining future position information. 10A is a conceptual diagram for explaining time synchronization and position matching, and FIG. 10B is a conceptual diagram showing an example of data processing based on identification. 10 is a flowchart illustrating a series of operations in the information providing system. 1 is a conceptual diagram showing an overview of an identification device and an information providing system. FIG. 10 is a conceptual diagram for explaining an outline of an identification device and an information providing system according to a second embodiment.

First Embodiment
An example of an identification device and an information provision system according to the first embodiment will be described below with reference to FIG. 1 and other figures. FIG. 1 is a conceptual diagram showing a crosswalk CW and its surroundings as a predetermined traffic area in which an identification device IE according to the present embodiment and an information provision system 100 including the same are installed. In the illustrated example, the crosswalk CW is located in an area without traffic lights. FIG. 2 is a diagram corresponding to FIG. 1 and is a conceptual diagram showing detection of a monitoring area DD by the identification device IE or the information provision system 100. FIG. 3 is a block diagram showing an example of the overall configuration of the information provision system 100, and FIG. 4 is a block diagram showing an example of the detailed configuration of the information provision system 100.

As shown in Figures 1 and 2, the identification device IE and information provision system 100 according to this embodiment are roadside installations that continuously communicate with the autonomous vehicle VE to understand the behavior of the autonomous vehicle VE (the planned future driving pattern of the autonomous vehicle VE, hatched with solid lines in Figure 1, is shown with dashed lines), detect moving objects MB that are present in a monitoring area DD that includes the crosswalk CW and its surrounding area, make collision predictions based on the detection results, and provide the autonomous vehicle VE with various information according to the prediction results.

Here, when performing collision predictions such as those described above, it is assumed that the infrastructure (roadside) has distinguished between the location of the autonomous vehicle VE itself (its own position) and other moving bodies (other vehicles, etc.). However, for example, as shown in the figure, in cases where another moving body MB, a vehicle GM, is traveling alongside the autonomous vehicle VE or there is a vehicle GM following the autonomous vehicle VE, it is not necessarily possible for the autonomous vehicle VE to be correctly identified on the roadside, and there is a risk of misrecognition.

In this embodiment, the identification device IE or the information provision system 100 equipped with it performs time synchronization on the information communicated between the identification device IE side (i.e., the roadside) and the autonomously driven vehicle VE side (i.e., the vehicle side), and then confirms the vehicle's position, thereby enabling the autonomously driven vehicle VE to be reliably identified (distinguished) from other objects. In particular, in this embodiment, time synchronization is performed based on satellite time information using GNSS (Global Navigation Satellite System) satellites ST, enabling more accurate self- and other-object identification.

To enable the above-described identification and the provision of information based on the identification results, the information provision system 100 is equipped with roadside equipment consisting of a driving assistance device SS, which serves as the main unit, and a sensor unit 10, which serves as the monitoring unit. A portion of these functions as the identification device IE for performing the above-described identification. The driving assistance device SS is equipped with a communication unit 30, a main control unit 50, and a GNSS receiver (GPS receiver) REs. As described above, the information provision system 100 is composed of roadside equipment installed at and around the crosswalk CW.

In the above configuration, as illustrated in FIG. 2, the information provision system 100 monitors a monitoring area (detection range) DD that includes the crosswalk CW and its surrounding area via the sensor unit 10 serving as a monitoring unit, and acquires target information that indicates the detection of a moving object MB (including an autonomous vehicle VE) present in the monitoring area (detection range) DD. In other words, when an object (moving object MB) present in the monitoring area DD is detected, the target information becomes object position information that indicates the location of the object.

In this case, the driving assistance device SS and the sensor unit 10, or a part of them, can also be said to function as an object information acquisition unit MR that detects objects present in the monitoring area DD and acquires target information as object position information indicating the location of the objects. Furthermore, it can also be said that the object information acquisition unit MR is equipped with the sensor unit 10 as a unit that senses the monitoring area DD and acquires object position information (target information), and extracts moving bodies MB present in the monitoring area DD based on the object position information (target information) from the sensor unit 10.

On the other hand, as illustrated in FIG. 1, the information provision system 100 communicates with an autonomously driven vehicle VE that is scheduled to pass through the crosswalk CW via the communication unit 30 to obtain future position information (information indicating its own future position) for the autonomously driven vehicle VE. The future position information presupposes information about the current position of the autonomously driven vehicle VE and is configured based on this current position. In other words, the future position information is information transmitted from the autonomously driven vehicle VE, and serves as self-position information indicating the autonomously driven vehicle VE's own position. In this case, the communication unit 30 accepts the future position information transmitted from the autonomously driven vehicle VE as self-position information, and functions as a vehicle information acceptance unit VR.

As described above, the main control unit 50 of the driving assistance device SS grasps the crosswalk CW and its surrounding conditions, as well as the behavior of the autonomously driven vehicle VE. Furthermore, based on the various pieces of information acquired above, i.e., the conditions of the crosswalk CW and its surroundings, and the behavior of the autonomously driven vehicle VE (travel plans determined from future position information), the main control unit 50 provides the autonomously driven vehicle VE with information such as the location of the moving object MB and the result of a determination as to whether the autonomously driven vehicle VE should temporarily stop in front of the crosswalk CW. For example, if the main control unit 50 determines that the autonomously driven vehicle VE should temporarily stop, it transmits information indicating the position of the virtual stop line VL to the autonomously driven vehicle VE, as illustrated in FIG. 1. From another perspective, the main control unit 50 functions as an information processing unit IP that processes various information and is one of the components of the information provision system 100.

To perform the above-described determinations, it is important to accurately grasp the relative positional relationship between the autonomous vehicle VE and the other moving body MB, i.e., to properly distinguish between the autonomous vehicle and the other moving body MB. Therefore, in this embodiment, as shown in the block diagram of an example configuration in Figure 3, both the information provision system 100 and the autonomous vehicle VE use GNSS as an absolute time reference for time synchronization. Specifically, in the information provision system 100, the driving assistance device SS links the target information (object position information) acquired by the sensor unit 10 with time information using GNSS satellites ST and records and manages the information. Meanwhile, the autonomous vehicle VE transmits its future position information (self-position information) to the information provision system 100, including time information using GNSS satellites ST. In short, the autonomous vehicle VE generates (acquires) the position and time information constituting its current position information, which is the basis for calculating its future position information, based on GNSS. The driving assistance device SS synchronizes the time between the time information contained in the future position information from the autonomously driven vehicle VE and the time information linked to the target information, and then compares the spatial information (e.g., coordinate data) contained in both to identify the position of the autonomously driven vehicle VE in the target information. This makes it possible to accurately identify the autonomously driven vehicle VE from among these, even when multiple vehicles are detected in close proximity, for example.

As mentioned above, in FIG. 1 , the autonomous vehicle VE is scheduled to cross the crosswalk CW, with the hatched position being its current position and its subsequent behavior (future planned operation) being indicated by the dashed line. That is, the autonomous vehicle VE transmits initial future position information to the driving assistance device SS, indicating that its future plans are as shown by the dashed line at its current position shown by the hatched position. This triggers communication between the autonomous vehicle VE and the driving assistance device SS. The autonomous vehicle VE continues to update and transmit its future position information, for example, at regular intervals. That is, the infrastructure side continuously receives new future position information from the autonomous vehicle VE. This communication continues, for example, until the autonomous vehicle VE has completely crossed the crosswalk CW or has passed through the monitoring area DD.

The driving assistance device SS takes into account continuously updated future position information to confirm whether the crosswalk CW, the destination of the autonomous vehicle VE, and the conditions around the crosswalk CW allow for continued driving, and provides the confirmation results to the autonomous vehicle VE. In other words, the driving assistance device SS can be considered an I2V (Infrastructure to Vehicle) control module IM that provides information from the infrastructure to the vehicle. In the illustrated example, the driving assistance device SS determines, based on the detection results of the sensor unit 10 (see Figure 2), that the autonomous vehicle VE needs to stop just before the crosswalk CW, and transmits position information of the virtual stop line VL, i.e., information on where the autonomous vehicle VE should stop.

The virtual stop line VL is a line that indicates a position set by the infrastructure, which is the information provider, but is not actually drawn on the road surface; it is position data. The position data (position information) of the virtual stop line VL is provided to the autonomously driven vehicle VE from the roadside as needed. For example, it is conceivable that the infrastructure could set the position of the virtual stop line VL as needed, and then transmit the position information of the set virtual stop line VL to the autonomously driven vehicle VE.

Below, a more specific example configuration of the information provision system 100 or the identification device IE constituting the information provision system 100 will be described with reference to the block diagram shown in Figure 4. As shown, and as previously described, the information provision system 100 includes a driving assistance device SS and a sensor unit 10. The driving assistance device SS includes a communication unit 30, a main control unit 50, and a GNSS receiver (GPS receiver) REs. Furthermore, in the illustrated example, the main control unit 50 is composed of various circuit boards, a CPU, a storage device, etc., and functionally includes an identification unit 80 that performs image processing to identify vehicles, etc.; a judgment unit JU that uses various information, such as the results of image processing, to make various decisions, such as whether to stop the autonomous vehicle VE; and a sensor interface (information acquisition unit) SEi for connecting to the sensor unit 10. In the main control unit 50, various processes are performed in each of these units, and necessary information (data) is input and output appropriately between each unit.

First, the sensor unit 10 is a roadside sensor composed of a camera unit 11 and a distance measurement unit 12. It acquires various data for detecting moving objects (MB) and obstacles within a monitoring area (DD), which is a predetermined area to be monitored. Moving objects (MB) may include vehicles on the road, as well as bicycles, pedestrians, and other obstacles along the roadside or on the sidewalk. The camera unit (infrastructure camera) 11 captures and generates image data to monitor the monitoring area (DD), which includes the roads before and after the crosswalk (CW), as well as sidewalks extending along the roadside. The distance measurement unit 12 may employ, for example, LiDAR, a millimeter-wave sensor, or radar, and performs distance measurement to generate distance data, thereby enabling the location of the moving object (MB). While only one sensor unit 10 is shown in the figure, multiple cameras can be installed within the site to thoroughly monitor the crosswalk (CW) and its surroundings. Additionally, while the monitoring area DD is shown here as an example, if the monitoring area DD changes depending on the direction of travel of the autonomously driven vehicle VE to which information is being provided, it is possible to appropriately select the cameras to be used accordingly. Here, the detection results obtained by the sensor unit 10 and various information such as image data and ranging data regarding moving objects MB present in the monitoring area DD are referred to as target information. In other words, the target information includes information on the operating status of various vehicles, pedestrians, etc. present in the monitoring area DD, as well as the presence of obstacles, etc.

The GNSS receiver (GPS receiver) REs is capable of acquiring time information using GNSS satellites ST, and in particular, in this case, the sensor interface (information acquisition unit) SEi of the main control unit 50 (described below) adds the time information acquired by the GNSS receiver REs to the target information from the sensor unit 10.

The communication unit 30 is a radio unit for wireless communication with the autonomously driven vehicle VE. As described above, the autonomously driven vehicle VE, which is the communication partner, transmits future position information indicating its own future position to the driving assistance device SS as data for receiving driving assistance. More specifically, the autonomously driven vehicle VE first generates future position information, consisting of information about the autonomously driven vehicle VE's current position and its planned future route based on the current position, in the future position information generation unit FG of the autonomous driving program AO to perform various controls for autonomous driving. This future position information includes the autonomously driven vehicle VE's current position (position at the current time) and its future position (including the predicted arrival time) created based on the current position, as well as information such as its speed and direction (azimuth angle) at each of these times (planned times). Therefore, by receiving the future position information from the autonomously driven vehicle VE, the driving assistance device SS can determine, for example, the predicted arrival time of the autonomously driven vehicle VE at the crosswalk CW and the time required to cross the crosswalk CW. In particular, in this embodiment, the autonomously driven vehicle VE has a GNSS receiver (GPS receiver) REv so that it can acquire current position information and time information using GNSS satellites ST. In other words, the current position (position at the current time) acquired by the GNSS receiver REv is used when generating future position information.

The sensor interface (information acquisition unit) SEi of the main control unit 50 acquires information acquired by the sensor unit 10, which is a monitoring unit, i.e., target information, and outputs it to the identification unit 80. That is, the information acquisition unit SEi acquires target information for a monitoring area DD, which is a predetermined range. In particular, in this embodiment, the information acquisition unit SEi also functions as a timestamp TS, and assigns time information using the GNSS receiver REs to the acquired target information. For example, as shown in parentheses in the figure, time information conforming to the GNSS is assigned in the timestamp TS to image data as target information as the acquisition time _tk (k≧1) of the image data, and the time-stamped image data is output as target information to the identification unit 80. Note that the image data may be, for example, multiple consecutively captured images, and the sequential acquisition times _tk (k≧1; tk ₊₁ ≧ _tk ) are numbered for these (see FIG. 5, described later).

Of the main control unit 50, the identification unit 80 has an image analysis unit 81, an analysis data storage unit 82, a matching unit 83, and an identification unit 84. Furthermore, the matching unit 83 is composed of a time synchronization unit 83a and a position matching unit 83b.

The image analysis unit 81 performs image analysis of target information composed of image data, etc. Specifically, it extracts partial images (object images) of moving objects MB, such as vehicles and people, included in the image data, classifies them by type, and extracts the position (coordinate data), shape, color, etc. of each moving object MB as needed. It also extracts the moving speed (velocity) from the displacement of the same moving object MB between consecutive images acquired as image data. The various data (analysis data) obtained as a result of the above analysis are stored in the analysis data storage unit 82. In particular, as described above, time information is added by the information acquisition unit SEi, and this time information is stored together with the analysis results as information accompanying the corresponding target information (image data). As described above, the image analysis unit 81, as the object information acquisition unit MR, extracts moving objects MB present in the monitoring area DD based on object position information (target information) from the sensor unit 10.

Of the collation unit 83, the time synchronization unit 83a performs time synchronization between roadside information and vehicle-side information. More specifically, time synchronization is performed based on the time information assigned to each piece of target information (image data) stored in the analysis data storage unit 82 and the current time included in the future position information acquired via the communication unit 30. In this case, the object information acquisition unit MR (driving assistance device SS) and the autonomously driving vehicle VE each have GNSS receivers REs and REv, allowing the collation unit 83 to perform time synchronization based on the satellite time information included in the target information (object position information) and future position information (self-position information).

Of the matching unit 83, the position matching unit 83b compares the position information (coordinate data) of the time-synchronized target object information (object position information) and future position information (self-position information), thereby identifying the autonomously driven vehicle VE from the target object information (object position information). In other words, the matching unit 83 compares the spatial coordinates of the time-synchronized target object information (object position information) and future position information (self-position information) to identify the location of the autonomously driven vehicle VE in the monitoring area DD.

The identification unit 84 identifies the autonomously driven vehicle VE from other objects (moving objects MB) in the monitoring area DD based on the matching results obtained by the matching unit 83 as described above. In essence, the identification unit 84 identifies the autonomously driven vehicle VE from among the moving objects MB extracted from image data by the image analysis unit 81 that constitutes the object information acquisition unit MR.

The judgment unit JU of the main control unit 50 has a stop judgment unit 52a that determines whether to issue a stop command to the autonomously driven vehicle VE in the notification unit 70, and a calculation unit 52b. Furthermore, the stop judgment unit 52a is composed of a pedestrian judgment unit JDp and a stop command unit SIs.

The pedestrian determination unit JDp of the stop determination unit 52a determines the presence or absence of a pedestrian CR at the crosswalk CW, i.e., the presence or absence of a pedestrian PE who is about to cross the crosswalk CW or has already started crossing, based on target information detected by the sensor unit 10. For example, if the sensor unit 10 detects a person who has remained within a predetermined range (e.g., within 2 to 3 meters) of a predetermined crosswalk CW for a predetermined period of time (e.g., several seconds), the pedestrian is determined to be a pedestrian CR. If the presence of such a pedestrian CR is confirmed and the future position information from the autonomously driven vehicle VE heading toward the crosswalk CW includes information indicating that the autonomously driven vehicle VE plans to cross the crosswalk CW, the stop command unit SIs of the stop determination unit 52a transmits a signal via the communication unit 30 to the autonomously driven vehicle VE to recommend or command it to stop (temporarily stop) at a predetermined virtual stop line VL. In addition to making a determination based on the presence or absence of a pedestrian CR as in the above example, the stop determination unit 52a also performs similar determination processing to the above in order to avoid collisions between various objects and the autonomously driven vehicle VE. In other words, the stop determination unit 52a functions as a determination unit that makes a predictive collision determination in order to avoid a collision.

The calculation unit 52b calculates the possible departure time (possible departure time) at which the autonomously driven vehicle VE, which is stopped at the virtual stop line VL, can depart from the virtual stop line VL. The possible departure time can typically be calculated by confirming, as a result of monitoring the monitoring area DD, that there is no moving body MB attempting to cross the road at the crosswalk CW or in its vicinity, or that they will no longer be present. It is anticipated that the possible departure time may be expressed, for example, as a set time (from what hour, minute, second to what hour, minute, second the vehicle can depart) or as a length of time (from what number of seconds after the present the vehicle can depart).

The communication unit 30 transmits information about the possible departure time calculated by the calculation unit 52b to the autonomously driven vehicle VE in the manner described above.

An example of the analysis of the target information will be described below with reference to FIG. 5. FIGS. 5A and 5B are data diagrams showing an example of an outline of the target information as a detection result of the monitoring area DD. In the example, the target information shows the results of analyzing image data in the image analysis unit 81 and storing them in the analysis data storage unit 82. FIG. 5A shows a list of the processing results of each image data stored in the analysis data storage unit 82 as a data diagram. That is, in the diagram, each image data D _k (k≧1) represents image data (continuous images) received via the information acquisition unit SEi, and analysis data resulting from the extraction of moving objects, etc. from each image data D _k by the image analysis unit 81 is linked to an acquisition time t _k (k≧1) indicating time information assigned by the time stamp TS. Details of each analysis data are composed of data related to the type and location information of the moving object MB, as exemplified in FIG. 5B, and are stored in the analysis data storage unit 82. The position coordinates, which are one of the parameters included in the position information, can be expressed, for example, as point coordinates in three-dimensional space. For example, a point corresponding to a predetermined reference point within a rectangular frame of an image of a vehicle cut out from image data can be used as a representative point indicating the position of the vehicle. In addition, the position information also includes information such as speed.

On the other hand, Figures 6(A) and 6(B) are data diagrams showing an example of an outline of the content of communication between the vehicle side and the roadside in the above-described configuration. Figure 6(A) shows information sent from the vehicle side to the roadside, and Figure 6(B) shows information sent from the roadside to the vehicle side, which are identified by ID. In other words, a judgment unit ID is defined for the roadside, and a vehicle ID is used for the vehicle side to identify the autonomously driven vehicle VE.

First, as shown in FIG. 6(A), the vehicle transmits various IDs, the data creation date and time T, and the location information (current location) and future location information of the autonomous vehicle VE to the roadside. The driving assistance device SS installed on the roadside receives this information. Here, the data creation date and time T and the location information (current location) of the autonomous vehicle VE are information obtained using the GNSS receiver (GPS receiver) REv, as described above. The location information (current location) includes information on the autonomous vehicle VE's speed (traveling speed) and direction (azimuth angle). In contrast, the future location information includes the same information as the location information (current location), but also includes information on the offset (distance) from the location information (current location). The future location information includes multiple (n) predicted values at regular intervals (e.g., every second) from the current time. In other words, roadside equipment can determine the planned route of the autonomous vehicle VE for, for example, n seconds into the future. With the above configuration, each piece of information is generated starting from location information (current location), and in this embodiment, this information is based on information from the GNSS receiver (GPS receiver) REv, resulting in information with high position accuracy over time.

On the other hand, as shown in FIG. 6(B), the roadside, i.e., the driving assistance device SS, transmits to the vehicle various IDs, creation dates and times, as well as information on the virtual stop line VL and the possible departure time (time of departure). In the illustrated example, the virtual stop line VL is provided with coordinates (latitude and longitude) of the start and end points, indicating its position as a line (line segment). Regarding the possible departure time, it is conceivable to provide literal time information. However, for example, transmitting a "possible departure" signal to the autonomously driven vehicle VE when it is possible to depart can also be considered as providing information equivalent to the possible departure time. Furthermore, in the illustrated example, target information is transmitted from the driving assistance device SS to the vehicle. In this case, the target information may exclude the position information of the autonomously driven vehicle VE itself and include information only about other moving objects (vehicles, etc.).

Future position information will be conceptually explained below with reference to Figure 7. Of Figures 7(A) to 7(D), Figure 7(A) first shows the initial future position information transmitted from autonomous vehicle VE to driving assistance system SS. The position where autonomous vehicle VE is depicted is the current position, and points FP1 to FP4 shown along the Z direction indicate the future positions of autonomous vehicle VE. More specifically, with time at the current position (current time) set to 0, point FP1 indicates the position of autonomous vehicle VE t seconds from the current time. Similarly, point FP2 indicates the position of autonomous vehicle VE 2t seconds from the current time, point FP3 indicates the position of autonomous vehicle VE 3t seconds from the current time, and point FP4 indicates the position of autonomous vehicle VE 4t seconds from the current time.

Next, Figure 7(B) shows the second future position information transmitted from autonomous vehicle VE to driving assistance system SS t seconds after the state of Figure 7(A). In this case, since t seconds have passed since the state of Figure 7(A), autonomous vehicle VE has traveled to a position corresponding to point FP1 in Figure 7(A). Then, autonomous vehicle VE transmits its new future position information, i.e., information on new points FP1 to FP4, to driving assistance system SS. Similarly, as shown in Figure 7(C), a third future position information (information on points FP1 to FP4) is transmitted another t seconds later.

On the other hand, if information about a virtual stop line VL is transmitted from the driving assistance device SS while the autonomously driven vehicle VE is continuing to drive, the autonomously driven vehicle VE will change its driving state during autonomous driving so as to decelerate and stop at the position of the corresponding virtual stop line VL. In this case, the future position information will also be changed, as shown in an example in Figure 7 (D).

In the above cases, the distance between each of points FP1 to FP4 indicates not only the physical distance traveled, but also the change in speed. Furthermore, the line (trajectory) connecting each point indicates the route and direction (azimuth angle) of the autonomous vehicle VE.

Note that the illustration only shows the case up to point FP4, and points beyond FP4 are omitted, but it is also possible to include information about future positions further in the future.

In the main control unit 50 of the driving assistance system SS, the judgment unit JU predicts collisions by taking into account information about the planned driving route of the autonomously driven vehicle VE as described above, as well as information about the crosswalk CW and its surroundings.

Hereinafter, time synchronization, position matching, etc. in the matching unit 83 (see FIG. 4) will be described with reference to the conceptual diagram shown in FIG. 8. For example, as shown in FIG. 8(A), an acquisition time _tk corresponding to the value of the data creation date and time T (also simply referred to as time T) in the future position information VD1 (partially excerpted from FIG. 6(A)) is selected from the data stored in the analysis data storage unit 82. Of the acquisition times _tk (k≧1), for example, data SSD1 (in the illustration, shown as acquisition time _tk ) that is closest to time T is selected as the corresponding data. In this manner, time synchronization is performed. Note that the number of selected data SSD1 is not limited to one, and two or more may be selected for one piece of future position information VD1.

Next, position matching is performed. Specifically, the analysis data in the data SSD1 is compared with the position information (current position) in the future position information VD1, and it is determined which data in the analysis data corresponds to the target autonomously driven vehicle VE, that is, the autonomously driven vehicle VE is identified. Specifically, as shown in FIG. 8(B), first, as shown in state α1, only vehicle objects indicated by hatching are selected from the analysis data of the data SSD1. Then, by comparing the position information (coordinate data) from these, one vehicle (vehicle B in the illustrated example) is extracted as corresponding to the autonomously driven vehicle VE, as shown in state α2. In this case, it is possible to set a certain tolerance range in the determination of consistency of the coordinate data, taking into account errors that may occur due to other factors, such as the difference between the time T and the acquisition time _tk and the magnitude of the speed, and to treat the data that meets the requirements (vehicle B in the illustrated example) as the autonomously driven vehicle VE. When an object that satisfies the requirements (vehicle B in the illustrated example) is determined, data from which the relevant data has been deleted is generated, as shown as state α3. In other words, data SSD2 is generated from information about the moving object MB as a result of monitoring the monitoring area DD, with information about the autonomously driven vehicle VE itself removed. This data SSD2 may be transmitted to the autonomously driven vehicle VE as data indicating the situation around the autonomously driven vehicle VE, or may be used within the driving assistance device SS as information for predicting a collision between the autonomously driven vehicle VE and another moving object MB. In other words, the autonomously driven vehicle VE is distinguished from other objects. In other words, the conceptual diagram shown in FIG. 8(B) illustrates an example of data diagram processing based on the distinction.

An example of a series of operations in the information provision system 100 will be described below with reference to the flowchart shown in Figure 9. Note that in the information provision system 100, monitoring of the monitoring area DD can be performed in various ways, but for example, such monitoring can be performed continuously, and target information can be accumulated continuously as needed.

First, in the driving assistance device SS, which is a roadside device, the main control unit 50 continues to check whether the first future position information, which serves as a trigger for starting communication, has been received (acquired) from the autonomously driven vehicle VE that is to receive driving assistance, until confirmation is obtained (step S101: Yes).

In step S101, when acquisition of the first future position information is confirmed (step S101: Yes), the main control unit 50 acquires target information (image data) (step S102). Note that while target information acquisition may begin in response to the above-mentioned trigger, it is also possible to select target information appropriately from target information that is continuously acquired, as described above.

Next, the main control unit 50, as the identification unit 80, performs time synchronization on the information acquired in steps S101 and S102, and at this time selects one or more corresponding data from the target information (step S103). After time synchronization in step S103, the selected data is compared with the position coordinates in the future position information (step S104).

Based on the processing in step S104, the main control unit 50 identifies the autonomously driven vehicle VE to be targeted from the target information (step S105), excludes information corresponding to the autonomously driven vehicle VE from the target information (step S106), and transmits a collision determination result based on the target information from which the information corresponding to the autonomously driven vehicle VE has been excluded to the autonomously driven vehicle VE (step S107).

Then, the main control unit 50 checks whether the autonomous vehicle VE has passed through the monitoring area DD (step S108). If passage is not confirmed (step S108: No), the main control unit 50 continues the confirmation operation for obtaining the next future position information until confirmation is confirmed (step S109: Yes).

In step S109, if it is confirmed that the next future position information has been acquired (step S109: Yes), the main control unit 50 repeats the operations from step S102. As a result, the main control unit 50 restarts by waiting for the next future position information to be acquired, and continues the above series of operations until the autonomously driven vehicle VE finally passes through the monitoring area DD (step S108: Yes).

On the other hand, if it is confirmed in step S108 that the autonomously driven vehicle VE has passed through the monitoring area DD (step S108: Yes), the main control unit 50 ends the series of operations and returns to the operations from step S101, i.e., begins detecting a new vehicle.

In addition, in step S107, the target information from which information corresponding to the autonomous vehicle VE has been excluded, i.e., the matched target information, may also be transmitted to the autonomous vehicle VE along with the collision determination result.

As a variation of the above, if the target autonomous vehicle VE cannot be identified from the target information in step S105, the system may wait until the next future position information is acquired and then identify the autonomous vehicle VE again.

Below, an overview of the identification device IE and information provision system 100 will be explained with reference to the conceptual diagram shown in Figure 10. Here, the identification device IE, which performs the above-mentioned identification, can be considered to be made up of a part of the information provision system 100, but from the functional perspective of enabling identification, for example, as shown in the figure, the identification device IE can be considered to be made up of an object information acquisition unit MR, a vehicle information reception unit VR, a matching unit 83, and an identification unit 84.

First, the object information acquisition unit MR detects objects present in the monitoring area DD and acquires object position information indicating their location. In other words, the object information acquisition unit MR detects vehicles (general vehicles) GM, autonomous vehicles VE, or other objects present in the monitoring area DD. Note that the object information acquisition unit MR can be considered to be composed of, in addition to the sensor unit 10, the information acquisition unit SEi and image analysis unit 81 of the driving assistance device SS that handle target information as object position information, as well as the GNSS receiver REs, etc.

Meanwhile, the vehicle information receiving unit VR receives self-location information transmitted from the vehicle that is to receive the information. In the example above, future location information as self-location information is received from the autonomously driving vehicle VE, which is the vehicle that is to receive the information, and the communication unit 30 functions as the vehicle information receiving unit VR.

Next, the matching unit 83 time-synchronizes the object position information (target information) acquired by the object information acquisition unit MR with the vehicle's own position information (future position information) received by the vehicle information reception unit VR, and matches the positions. In particular, in this embodiment, time information conforming to GNSS is used as an absolute standard for time synchronization, enabling highly accurate synchronization between the road and the vehicle before position matching.

Finally, the identification unit 84 identifies the autonomously driven vehicle VE from other objects in the monitoring area DD based on the matching results from the matching unit 83.

As described above, the identification device IE of this embodiment comprises an object information acquisition unit MR that detects objects present in the monitoring area DD and acquires object position information indicating their location; a vehicle information reception unit VR that receives self-position information transmitted from a vehicle (autonomously driven vehicle VE); a matching unit 83 that time-synchronizes the object position information with the self-position information and compares their positions; and an identification unit 84 that distinguishes the autonomously driven vehicle VE from other objects in the monitoring area DD based on the matching results of the matching unit 83. The identification device IE time-synchronizes the object position information for the monitoring area DD with the self-position information from the vehicle (autonomously driven vehicle VE) and then performs position matching between them. In other words, by time-synchronizing information from the vehicle and information from the roadside and then performing position matching, it is possible to reliably identify the vehicle (autonomously driven vehicle VE) from other objects in the monitoring area DD.

The information provision system 100 according to this embodiment also includes the above-mentioned identification device IE, and transmits the analysis results of the object position information, including the identification by the identification unit 84, to the autonomously driven vehicle VE. By including the identification device IE, the information provision system 100 can reliably identify the vehicle (autonomously driven vehicle VE) from other objects in the monitoring area DD, and can transmit information about the identification results to the vehicle (autonomously driven vehicle VE).

Second Embodiment
An example of the information provision system according to the second embodiment will be described below with reference to Fig. 11. Fig. 11 is a conceptual diagram for outlining an example of operation at an intersection CS where the information provision system 100 according to this embodiment is installed, and corresponds to Fig. 1 and the like.

In the first embodiment, no particular mention was made of information provision (driving assistance) when the information provision system 100 is installed at an intersection CS. In contrast, in this embodiment, as an example, a case where an autonomous vehicle VE is about to turn right at an intersection CS is shown. Note that, apart from the above points, the present embodiment is the same as the first embodiment, so explanations of the overall configuration of the information provision system 100 will be omitted, and other figures, such as those corresponding to those described above, will be used as appropriate as necessary.

To further explain the illustrated embodiment, first, an autonomously driven vehicle VE, shown hatched, is about to make a right turn at an intersection CS, as indicated by arrow AA1. Meanwhile, an oncoming vehicle GM is approaching the same intersection CS from the oncoming lane in which the autonomously driven vehicle VE is traveling, as indicated by open arrow BB1. If the oncoming vehicle GM plans to continue straight ahead, the autonomously driven vehicle VE will need to stop (temporarily stop) at the intersection CS. In response, the information provision system 100 pre-sets a virtual stop line VL just before the expected passage area IL through which the oncoming vehicle GM is expected to pass, and transmits to the autonomously driven vehicle VE position information of the set virtual stop line VL, i.e., information regarding where the autonomously driven vehicle VE should stop. Again, based on the identification result by the identification unit 84 (i.e., assuming the use of the identification result), a collision determination between the autonomously driven vehicle VE and the oncoming vehicle GM, which is a moving body other than the autonomously driven vehicle VE, is performed by the determination unit JU, which serves as a determination unit.

In this embodiment, too, it is possible to reliably distinguish between a vehicle (autonomously driven vehicle VE) and other objects in the monitoring area DD, and the position information of the virtual stop line VL as information on the results of the distinction can be transmitted to the vehicle (autonomously driven vehicle VE). In particular, this embodiment can improve safety at intersections CS.

〔others〕
The present invention is not limited to the above-described embodiment, and can be embodied in various forms without departing from the spirit and scope of the present invention.

First, the locations where the information provision system 100 can be introduced are not limited to those mentioned above, and the information provision system 100 can be applied to various locations where the autonomously driven vehicle VE requires information provision.

In addition, in the above, the target vehicle for information provision in the information provision system 100 is an autonomous vehicle VE, but vehicles other than autonomous vehicles can also be the target for receiving information (driving assistance) as long as they are capable of communicating with the information provision system 100 and capable of highly accurate synchronization.

In addition, while the above describes the use of GNSS as a means of time synchronization, other methods can also be used as long as they are capable of achieving the required level of accuracy. Furthermore, while the above uses GNSS to perform synchronization using it as an absolute standard, it is also possible to use a form in which synchronization is performed relatively between the vehicle side and the roadside, for example. It is also possible to combine and apply multiple methods. Alternatively, the method applied may change depending on the installation location of the roadside devices that make up the information provision system 100.

In addition, while target information has been shown above as an example of object position information and future position information as an example of self-position information, various other aspects of object position information and self-position information are also conceivable. Regarding target information, while analysis of image data has been described as a representative example, it is conceivable that it can be used with various data that can extract and identify vehicles, etc., rather than just image data. For example, ranging data, among the above, can also be handled in the same way as image data. Furthermore, although image analysis processing is described above as being performed in the identification unit 80, various other aspects are possible, and analysis processing is not necessarily limited to all target information, and can be performed as needed.

In addition, while the above description has been given assuming that the information provision system 100 provides information (driving assistance) to a single autonomously driven vehicle VE that is the assistance target, if there are two or more assistance targets at the same time, it is also possible to respond to them simultaneously in parallel. In this case, for example, when predicting a collision, future position information from two or more autonomously driven vehicles VE can be taken into consideration comprehensively.

The reference position of the autonomously driven vehicle VE can also be set in various ways. For example, as described above, when the current position is obtained using the GNSS receiver REv, the obtained position will be the installation position of the GNSS receiver REv on the autonomously driven vehicle VE. In this case, it is possible to include information on the installation position of the GNSS receiver REv in the future position information, along with information on the shape and size of the autonomously driven vehicle VE. Alternatively, the information may be corrected to indicate the reference position of the autonomously driven vehicle VE, taking into account the installation position of the GNSS receiver REv, and then transmitted.

In addition, if there is a difference between the reference position of the autonomously driven vehicle VE set in the future position information and the reference position set for each moving body MB obtained as a result of analyzing image data determined by the information provision system 100, this difference may be taken into account when determining the tolerance for the consistency of the coordinate data.

Furthermore, even if future position information is transmitted but the autonomously driven vehicle VE that should be at that location is hidden behind another vehicle and cannot be detected by the sensor unit 10, it is possible to trust the future position information and perform processing if it is determined from the situation that there is space in the relevant location for the autonomously driven vehicle VE.

Furthermore, for example, the virtual stop line VL and the expected passing area IL may be further adjusted in terms of buffer amount, etc., taking into account weather conditions such as rain or snow. In other words, the position setting of the virtual stop line VL may be changed depending on the weather.

In addition, in the above description, the driving assistance device SS that constitutes the information provision system 100 is installed near the site, but this is not limited to this. For example, locations responsible for various information processing and data management could be installed in a remote location as a management center (management server), or various processing and data storage could be performed on the cloud.

In addition, in the above example, the sensor unit 10 is configured to include a camera unit 11 and a distance measurement unit 12, but the configuration of the sensor unit 10 is not limited to this, and the sensor unit 10 may be configured to include, for example, either the camera unit 11 or the distance measurement unit 12.

10...sensor unit, 11...camera unit, 12...distance measurement unit, 30...communication unit, 50...main control unit, 52a...stop determination unit, 52b...calculation unit, 70...alert unit, 80...identification unit, 81...image analysis unit, 82...analysis data storage unit, 83...matching unit, 83a...time synchronization unit, 83b...position matching unit, 84...identification unit, 100...information provision system, AA1...arrow, AO...autonomous driving program, BB1...arrow, CR...pedestrian, CS...intersection, CW...pedestrian crossing, DD...monitoring area, Dk...image data, FG...future position information generation unit, FP1 to FP4...point, GM...vehicle (general vehicle, oncoming vehicle), IE...identification device, IL...expected passage area, IM...I2V control module, IP...information processing unit, JDp...pedestrian detection unit, JU...detection unit (detection unit), MB...moving object, MR...object information acquisition unit, PE...pedestrian, REs...GNSS receiver (GPS receiver), REv...GNSS receiver (GPS receiver), SEi...sensor interface (information acquisition unit), SIs...stop command unit, SS...driving assistance device, SSD1, SSD2...data, ST...GNSS satellite, T...data creation date and time (time), TS...timestamp, VD1...future position information, VE...autonomous vehicle, VL...virtual stop line, VR...vehicle information reception unit, tk...acquisition time, α1-α3...state

Claims (7)

an object information acquisition unit that detects an object present in a monitoring area and acquires object position information indicating the object's position in association with satellite time information ;
a vehicle information receiving unit that continuously communicates with a target vehicle in the monitoring area and receives self-position information including satellite time information transmitted from the target vehicle;
a matching unit that time-synchronizes the object position information and the self-position information using satellite time information and matches the positions;
An identification device comprising an identification unit that identifies the target vehicle from among a plurality of vehicles present in the monitoring area based on the matching result by the matching unit. The identification device according to claim 1 , wherein the matching unit compares spatial coordinates in the object position information and the self-position information, which are time-synchronized, to identify the location of the target vehicle in the monitoring area. The object information acquisition unit and the target vehicle each have a GNSS receiver,
The identification device according to claim 1 , wherein the matching unit performs time synchronization based on satellite time information included in the object location information and the self-location information. the target vehicle is an autonomous vehicle,
The identification device according to any one of claims 1 to 3, wherein the vehicle information accepting unit accepts future position information of the autonomously driven vehicle as the self-position information. the object information acquisition unit includes a sensor unit that senses the monitoring area and acquires the object position information, and extracts a moving object that exists in the monitoring area based on the object position information from the sensor unit;
5. The identification device according to claim 1, wherein the identification unit identifies the target vehicle from among the moving objects extracted by the object information acquisition unit. The identification device according to claim 5 , further comprising a determination unit that determines a collision between the target vehicle and a moving object other than the target vehicle based on the identification result by the identification unit. The identification device according to any one of claims 1 to 6 is provided,
an information provision system that transmits an analysis result regarding the object position information, including the identification by the identification unit, to the target vehicle;