JP2020154619A - Vehicle controller, object monitoring device, and vehicle control system - Google Patents

Abstract

To mitigate the influence of a detection result of both an in-vehicle sensor and a road sensor being mixed for the same object when the in-vehicle sensor and the road sensor cooperatively detect an object around a vehicle.SOLUTION: A vehicle controller according to the present invention receives obstacle data describing a position of a first object and a position of a second object from an object monitoring device installed outside a vehicle, and regards the second object as a blind spot obstacle when the first object is the vehicle.SELECTED DRAWING: Figure 6B

Description

The present invention relates to a vehicle control device that controls a vehicle.

As a technology for supporting the driving of a vehicle, there is a technology for notifying the driver of an object detected by a sensor mounted on the vehicle or a sensor installed on the road. In-vehicle sensors and road sensors cannot always detect all objects around the vehicle, and objects may exist in areas that are blind spots for the vehicle. Several techniques have been developed on how to notify the driver of such blind spot obstacles.

The following Patent Document 1 "provides a vehicle system, a vehicle information processing method, a program, and a traffic system capable of safely traveling a vehicle. "The vehicle system 50 acquires the first blind spot information indicating the blind spot region generated by the object when viewed from the infrastructure sensor 31 that detects surrounding objects from the infrastructure system 3 having the infrastructure sensor 31." The acquisition unit is provided with a detection / recognition unit 52 that generates second blind spot information indicating a blind spot region generated by the object when viewed from the vehicle-mounted sensor 51 that detects the object. Then, the detection recognition unit 52 has an integrated unit that outputs common blind spot information indicating a first common blind spot area common to each blind spot area based on the first blind spot information and the second blind spot information to an external device. 』(See summary).

The following Patent Document 2 provides, "Providing a driving support device capable of calculating a blind spot region created by an object and making it difficult for moving objects existing in the blind spot region of the object to come into contact with each other. The subject is "the position of the blind spot object A ₁ that generates the blind spot areas F ₁ and F _2, and the moving objects B ₁ and B existing in the vicinity of the blind spot object A ₁ and the blind spot object A _1". based on the relative position of the _2, it calculates the blind spot object _a moving body _B 1, produced by _a blind spot region of the _{B 2 F} 1, _{F 2.} Thus, it is possible to recognize the blind area of the blind spot object _{A 1} mobile _B 1, _{B 2} by _F 1, _{F 2,} for example, the mobile _B 1, _{B 2} existing in the blind area _F 1, _{F 2} By calling attention by issuing an alarm, it is possible to make it difficult for moving objects B ₁ and B ₂ existing in different blind spot regions created by the blind spot object A ₁ to come into contact with each other. 』(See summary).

JP-A-2018-195289 Japanese Unexamined Patent Publication No. 2010-079655

When the in-vehicle sensor and the road sensor detect the same object, the object position detected by the in-vehicle sensor and the object position detected by the road sensor may deviate from each other. This is because the communication time and calculation time between the vehicle and the road sensor are required. At this time, it is difficult for the vehicle to determine whether two objects exist at that position or whether the same object is detected at a different position. On the other hand, such a judgment is not necessary for an object that cannot be detected by the in-vehicle sensor and is detected only by the road sensor. This is because the information about the position of the object is only obtained from the road sensor. In view of the above, it is desirable to mitigate the influence of the mixture of the detection results of both the in-vehicle sensor and the road sensor for the same object.

In the techniques described in Patent Documents 1 and 2, detection of an object existing in the blind spot region is considered, but as described above, the detection results of both the in-vehicle sensor and the road sensor are mixed for the same object. It is considered that no special consideration has been given to the effects of this.

The present invention has been made in view of the above problems, and when the in-vehicle sensor and the road sensor cooperate to detect an object around the vehicle, the detection results of both the in-vehicle sensor and the road sensor are obtained for the same object. The purpose is to mitigate the effects of mixing.

The vehicle control device according to the present invention receives obstacle data describing the position of the first object and the position of the second object from an object monitoring device installed outside the vehicle, and the first object is the vehicle. If there is, the second object is regarded as a blind spot obstacle.

According to the vehicle control device according to the present invention, if one of the objects described in the obstacle data transmitted by the object monitoring device is the own vehicle, the other is regarded as a blind spot obstacle, and thus is regarded as a blind spot obstacle. You can narrow down the objects. This makes it possible to mitigate the effect of mixing the detection results of both the in-vehicle sensor and the road sensor for the same object.

It is a block diagram of the vehicle control system 1 which concerns on Embodiment 1. It is a figure which shows typically the range of the object which can be detected by the vehicle control device 100. It is a figure which shows typically the object which the object monitoring apparatus 200 detected. It is a figure which shows the state when the vehicle control device 100 receives the object position detected by the object monitoring device 200. It is a functional block diagram of each device included in a vehicle control system 1. It is a flowchart explaining the operation procedure of the object detection unit 220. It is a conceptual diagram which shows how the object detection part 220 carries out S406. It is a flowchart explaining the operation procedure of the terrain acquisition part 230. It is a flowchart explaining the operation procedure of the blind spot relation extraction part 240. It is a conceptual diagram which shows how the blind spot relation extraction part 240 carries out S603. It is a flowchart explaining the operation procedure of the blind spot obstacle extraction unit 140. It is a conceptual diagram which shows how the blind spot obstacle extraction unit 140 carries out S704. It is a flowchart explaining the detailed procedure of S704. It is a flowchart explaining the operation procedure of the traveling control unit 150. This is an example of a grid that describes the coordinates around the object monitoring device 200. It is a functional block diagram of each apparatus in Embodiment 2. FIG. It is a schematic diagram which shows the result of the object detection unit 220 detecting an obstacle in Embodiment 2. FIG. 5 is a conceptual diagram showing how the blind spot relationship extraction unit 240 identifies an obstacle pair having a blind spot relationship in the second embodiment. It is a conceptual diagram which shows how the area data and obstacle data are transmitted and received between the object monitoring device 200 and the vehicle control device 100. It is a functional block diagram of the vehicle control device 100 which concerns on Embodiment 3.

<Embodiment 1>
FIG. 1 is a configuration diagram of a vehicle control system 1 according to a first embodiment of the present invention. The vehicle control system 1 is a system that controls the operation of the vehicle 10, and includes a vehicle control device 100 and an object monitoring device 200. The vehicle control device 100 is a device mounted on the vehicle 10 and electronically controls each functional unit included in the vehicle 10. The object monitoring device 200 is a device that is installed outside the vehicle 10 (for example, on the road) and detects an object existing in the vicinity of the object monitoring device 200.

FIG. 2A is a diagram schematically showing a range of objects that can be detected by the vehicle control device 100. In the scene shown in FIG. 1, a shield 30 (for example, a wall having a certain height) exists around the road on which the vehicle 10 is traveling. The pedestrians 21 and 22 are traveling in directions orthogonal to the traveling direction of the vehicle 10, respectively. The pedestrian 21 is present at a position that becomes a blind spot (hatched portion in FIG. 2A) for the vehicle control device 100 due to the shield 30. That is, the vehicle control device 100 cannot recognize the pedestrian 21. The pedestrian 22 is located at a position that can be recognized by the vehicle control device 100.

FIG. 2B is a diagram schematically showing an object detected by the object monitoring device 200. In the scene shown in FIG. 1, the object monitoring device 200 can detect the vehicle 10, the pedestrians 21 and 22, and the shield 30. Since the object monitoring device 200 can detect the pedestrian 21, there is a case where the vehicle control device 100 cannot recognize the pedestrian 21 by transmitting the position of the pedestrian 21 from the object monitoring device 200 to the vehicle control device 100. However, the vehicle control device 100 can grasp the position and take an avoidance action or the like.

FIG. 2C is a diagram showing a state when the vehicle control device 100 receives the object position detected by the object monitoring device 200. When the position of the pedestrian 21 detected by the object monitoring device 200 is transmitted from the object monitoring device 200 to the vehicle control device 100, walking is caused by the communication time between the object monitoring device 200 and the vehicle control device 100. There may be a gap between the actual position of the person 21 and the position recognized by the vehicle control device 100. For example, when the object monitoring device 200 recognizes the pedestrian 21 in FIG. 2C and transmits the position to the vehicle control device 100, the pedestrian 21 reaches the position 21a when the vehicle control device 100 receives the position. It may be moving. Similarly, the pedestrian 22 may have moved to the position 22a.

Since the vehicle control device 100 cannot recognize the pedestrian 21, even if the position of the pedestrian 21 deviates slightly from the position received from the object monitoring device 200, the position of the pedestrian 21 can be recognized as long as the presence of the pedestrian 21 can be recognized. It is useful to get. As for the pedestrian 22, the vehicle control device 100 recognizes the pedestrian 22 on the position 22a, whereas the object monitoring device 200 recognizes the pedestrian 22 on the solid line in FIG. 2C. The vehicle control device 100 acquires two positions of the pedestrian 22. Then, it is difficult for the vehicle control device 100 to determine whether there are two objects at the positions, or whether the vehicle control device 100 and the object monitoring device 200 recognize the same object with a deviation.

In view of the above problems, the first embodiment provides a method capable of notifying the vehicle control device 100 of the position of only an object that is a blind spot for the vehicle control device 100 from the object monitoring device 200. It was decided to. As a result, the vehicle control device 100 can acquire the position (although it may be slightly deviated) from the object monitoring device 200 for the pedestrian 21, and the vehicle control device 100 recognizes the pedestrian 22. Only the position can be obtained. Therefore, it is considered that the possibility that a plurality of detection results for the position of the same object are mixed on the vehicle control device 100 can be suppressed.

FIG. 3 is a functional block diagram of each device included in the vehicle control system 1. The object monitoring device 200 includes a peripheral monitoring unit 210, an object detection unit 220, a terrain acquisition unit 230, a blind spot relationship extraction unit 240, and an obstacle data transmission unit 250. The vehicle control device 100 includes a peripheral monitoring unit 110, an obstacle data receiving unit 120, an object detecting unit 130, a blind spot obstacle extracting unit 140, a traveling control unit 150, and a control executing unit 160.

The peripheral monitoring unit 210 detects the position of an object existing in the vicinity of the object monitoring device 200. For example, the peripheral monitoring unit 210 can be configured by an image sensor such as a camera or a distance measuring sensor such as LiDAR. When an image sensor is used, the peripheral monitoring unit 210 outputs a monitoring image around the object monitoring device 200. When the distance measuring sensor is used, the peripheral monitoring unit 210 outputs point cloud data representing the coordinates of the boundary portion of the object. Other suitable detection devices can also be used.

The object detection unit 220 creates a background map in which the coordinates of fixed objects around the object monitoring device 200 are described according to a flowchart described later, and detects obstacles around the object monitoring device 200. The details of the obstacle will be described later. The terrain acquisition unit 230 creates a shield map in which the coordinates of the shield 30 are described according to a flowchart described later. The blind spot relationship extraction unit 240 identifies a pair of two obstacles having a blind spot relationship according to a flowchart described later. Details of the blind spot relationship will be described later.

The obstacle data transmission unit 250 transmits obstacle data describing the positions of pairs having a blind spot relationship among the obstacles detected by the object monitoring device 200. By transmitting only the coordinates of the obstacle pair having a blind spot relationship, the problem described in FIG. 2C is solved. However, the obstacle data transmission unit 250 transmits the obstacle data by broadcast transmission without distinguishing whether or not the transmission destination is the vehicle 10. Therefore, the vehicle control device 100 needs to determine whether or not one of the obstacle pairs described in the obstacle data is the vehicle 10 itself by the procedure described later.

The peripheral monitoring unit 110 detects the position of an object existing around the vehicle 10. The peripheral monitoring unit 110 can be configured by the same device as the peripheral monitoring unit 210. The obstacle data receiving unit 120 receives obstacle data from the object monitoring device 200. As described above, this obstacle data may include the coordinates of an obstacle other than the vehicle 10. The object detection unit 130 detects obstacles existing around the vehicle 10 according to the detection result by the peripheral monitoring unit 110. The detection procedure may be the same as that of the object detection unit 130, or other procedures may be used. The blind spot obstacle extraction unit 140 identifies an obstacle described in the obstacle data that has a blind spot relationship with the vehicle control device 100 according to a flowchart described later. The travel control unit 150 and the control execution unit 160 control the operation of the vehicle 10 according to a flowchart described later.

FIG. 4A is a flowchart illustrating the operation procedure of the object detection unit 220. Each step of FIG. 4A will be described below. In the following, it is assumed that the peripheral monitoring unit 210 is composed of a distance measuring sensor.

(FIG. 4A: steps S401-S402)
The object detection unit 220 acquires three-dimensional point cloud data from the peripheral monitoring unit 210 (S401). The object detection unit 220 creates a two-dimensional map (XY map) according to the procedure illustrated below using the three-dimensional point cloud data (S402). The range and resolution of the XY map are determined in advance. For example, the range is ± 50 m in the X direction and ± 50 m in the Y direction centering on the peripheral monitoring unit 210, and the resolution is 20 cm per grid.

(FIG. 4A: Step S402: Example 1 of procedure for creating XY map)
The peripheral monitoring unit 210 may also detect the ground as a peripheral object. Therefore, in order to exclude the ground from the surrounding objects, the object detection unit 220 keeps the coordinates in the Z direction (vertical direction) within a predetermined range (for example, 30 cm or more and 2 m or less) among the XYZ coordinates described in the point cloud data. Only certain ones are adopted as XY coordinates on the XY map. The same effect can be obtained by the following examples 2 and 3.

(FIG. 4A: Step S402: Example of procedure for creating an XY map Part 2)
When the peripheral monitoring unit 210 is a distance measuring sensor, the following procedure can be used instead of Example 1. A coordinate point having a value in the Z direction (an object boundary exists) on the 3D point cloud data is projected on the XY grid. For coordinate points belonging to the same XY grid, the values in the Z direction are added. Only the XY grid whose Z coordinate value after addition is equal to or greater than a predetermined value is left on the XY map. As a result, only the XY coordinate points having a certain width in the Z direction are reflected on the XY map.

(FIG. 4A: Step S402: Example of procedure for creating XY map No. 3)
When the peripheral monitoring unit 210 is a distance measuring sensor, the following procedure can be used instead of Example 1 or Example 2. The reflection intensity values of LiDAR belonging to the same XY grid are added. Only the XY grid whose reflection intensity value after addition is equal to or more than a predetermined value is left on the XY map. As a result, only the XY coordinate points having a certain width in the Z direction are reflected on the XY map.

(FIG. 4A: Step S402: Supplement)
When the peripheral monitoring unit 210 is composed of a camera, the ground portion of the monitored image can be removed by using the camera installation position and shooting direction. The position of the peripheral object can be specified, for example, by detecting the boundary position on the surveillance image.

(FIG. 4A: steps S403 to S404)
The object detection unit 220 determines whether or not the background map is completed (S403). If it is completed, the process proceeds to step S405. If it is not completed, the background map is updated by applying the result of step S402 to the background map (S404). Returning to step S401, the point cloud data at a new time is acquired, and the same process is repeated.

(FIG. 4A: Steps S403 to S404: Supplement)
A background map can be created by averaging a predetermined number or more of the XY maps created in step S402. This is because it is considered that an XY map describing the XY coordinates of a fixed object can be obtained by averaging the XY maps over time. However, when there are many moving objects around the object monitoring device 200 or when there are many temporarily placed objects, it is considered that the accuracy of the XY coordinates of the fixed objects is low even if the XY maps are averaged. Therefore, in S403, the object detection unit 220 sets the background map when the integrated number of XY maps does not reach the predetermined number, or when the dispersion of the object positions described in the averaged XY map is equal to or more than the predetermined value. It is considered incomplete.

(FIG. 4A: steps S405-S406)
The object detection unit 220 creates a difference map by obtaining the difference between the XY map created in S402 and the completed background map (S405). The object detection unit 220 considers that a moving object or a temporary object exists on the XY grid existing on the difference map, and identifies these objects as obstacles (S406).

FIG. 4B is a conceptual diagram showing how the object detection unit 220 executes S406. Here, an example of detecting the boundary surface of a peripheral object as point cloud data in the situation of FIG. 1 is shown. The right figure of FIG. 4B is the point cloud data. FIG. 1 is transcribed in the left figure of FIG. 4B. The object detection unit 220 detects the boundary surface of an object other than the shield 30 as point data by creating a difference map in S405. The object detection unit 220 recognizes the range of each object by grouping the point data. For example, point clouds that are close to each other can be grouped as objects by regarding them as the same object. In the example of FIG. 4B, the object IDs 1 to 3 corresponding to the vehicle 10 / pedestrian 21 / pedestrian 22 were recognized.

FIG. 5 is a flowchart illustrating the operation procedure of the terrain acquisition unit 230. The terrain acquisition unit 230 implements this flowchart after the background map is completed. Each step of FIG. 5 will be described below.

(Fig. 5: Step S501)
The terrain acquisition unit 230 acquires three-dimensional point cloud data from the peripheral monitoring unit 210.

(Fig. 5: Step S502)
The terrain acquisition unit 230 creates a shield candidate map that describes the XY coordinates in which the shield candidate exists. Specifically, first, a highest point map describing the highest point (largest Z coordinate) at each XY coordinate and a lowest point map describing the lowest point (smallest Z coordinate) at each XY coordinate are created. The terrain acquisition unit 230 specifies the XY coordinates in which the difference between the highest point and the lowest point in the same XY coordinates is equal to or greater than a predetermined value. That is, the XY coordinates of an object having a certain width in the height direction are specified.

(Fig. 5: Step S503)
The terrain acquisition unit 230 identifies the XY grid in which an object that is stationary and has a certain width in the height direction exists by superimposing the XY coordinates specified in S502 and the background map. This XY grid can be regarded as the coordinates of the shield 30. The XY grid created by this step will be called a shield map.

(Fig. 5: Step S503: Supplement)
Since the peripheral monitoring unit 210 can acquire the three-dimensional coordinates of the peripheral object of the object monitoring device 200, the obstacle data transmission unit 250 can also transmit the obstacle data in which the three-dimensional coordinates are described. .. However, processing the three-dimensional coordinate data has a large calculation load, and it is sufficient if the vehicle control device 100 can acquire the two-dimensional coordinates in order to avoid a collision. Therefore, the obstacle data is described by two-dimensional coordinates, and the data in the height direction is used only when the candidate of the shield 30 is specified. This has the advantage that the communication load and calculation load related to obstacle data can be suppressed.

FIG. 6A is a flowchart illustrating the operation procedure of the blind spot relationship extraction unit 240. The blind spot relationship extraction unit 240 implements this flowchart after the shield map is completed. Each step of FIG. 6A will be described below.

(FIG. 6A: steps S601 to S602)
The blind spot relationship extraction unit 240 acquires the position of the obstacle specified by the object detection unit 220 (S601). The blind spot relationship extraction unit 240 selects any two of the obstacles and projects them onto the shield map (S602).

(FIG. 6A: steps S603 to S605)
The blind spot relationship extraction unit 240 determines whether or not there is a shield between the two obstacles selected in S602 (S603). The specific procedure of S603 will be described later. The blind spot relationship extraction unit 240 stores two obstacles having a shield between them as a pair of obstacles having a blind spot relationship (S604). The blind spot relationship extraction unit 240 carries out S602 to S604 for all combinations of obstacles (S605).

FIG. 6B is a conceptual diagram showing how the blind spot relationship extraction unit 240 executes S603. Here, the same example as in FIG. 4B is shown. The blind spot relationship extraction unit 240 connects each center of gravity of two obstacles with a line segment, and determines whether or not the two obstacles have a blind spot relationship depending on whether or not there is a shield on the line segment. To do. In the example of FIG. 6B, since the shield 30 exists between the object ID = 1 (vehicle 10) and the object ID = 2 (pedestrian 21), it can be seen that these objects are in a blind spot relationship.

FIG. 7 is a flowchart illustrating the operation procedure of the blind spot obstacle extraction unit 140. Each step of FIG. 7 will be described below.

(FIG. 7: Steps S701 to S702)
The blind spot obstacle extraction unit 140 acquires the position of the vehicle 10 from, for example, the GPS (Global Positioning System) system mounted on the vehicle 10 (S701). The blind spot obstacle extraction unit 140 acquires an arbitrary obstacle pair described in the obstacle data (S702).

(Fig. 7: Step S703)
The blind spot obstacle extraction unit 140 determines whether or not any one of the obstacle pairs selected in S702 is the vehicle 10. Specifically, it is determined whether or not the position of the vehicle 10 acquired in S701 and the position of any one of the obstacle pairs match (the difference between the two positions is within the determination threshold value). If any one of the obstacle pairs is the vehicle 10, the process proceeds to S704, and if none of the obstacle pairs is the vehicle 10, the process proceeds to S706.

(Fig. 7: Step S703: Supplement)
The object monitoring device 200 transmits the obstacle data by broadcast transmission without determining whether or not the destination of the obstacle data is the vehicle 10. Therefore, the vehicle 10 may receive, for example, an obstacle pair other than the object ID = 1 and the object ID = 2 in FIG. 6B. This step is for extracting an obstacle pair that describes an obstacle that has a blind spot relationship for the vehicle 10 from the obstacle data transmitted by the object monitoring device 200. This is because the obstacle data describes a pair of two objects having a blind spot relationship, and if one of the two objects is the vehicle 10, the other is a blind spot obstacle for the vehicle 10.

(FIG. 7: Steps S704 to S706)
The blind spot obstacle extraction unit 140 corrects the position of the obstacle acquired in S702 to S703 according to the flowchart described later (S704). The blind spot obstacle extraction unit 140 saves the corrected obstacle position (S705). The blind spot obstacle extraction unit 140 carries out S702 to S705 for all the combinations of obstacle pairs described in the obstacle data (S706).

FIG. 8A is a conceptual diagram showing how the blind spot obstacle extraction unit 140 executes S704. Here, it is assumed that the vehicle 10 is traveling straight to the right and the detection surface of the peripheral monitoring unit 210 is arranged on the extension of the traveling direction (FIG. 8A (1)). When the peripheral monitoring unit 210 is composed of a distance measuring sensor such as LiDAR, the peripheral monitoring unit 210 detects the position of a peripheral object based on the beam irradiation direction and the reflection intensity. Therefore, if the installation angle of the object monitoring device 200 deviates due to the influence of, for example, an environmental change, the detection result by the peripheral monitoring unit 210 increases the possibility that the object detection direction deviates (FIG. 8A (2)). Therefore, the blind spot obstacle extraction unit 140 corrects this directional deviation.

In the positional relationship shown in FIG. 8A, it is assumed that the directional deviation between the actual position of the vehicle 10 and the position of the vehicle 10 recognized by the object monitoring device 200 is the angle θ. The blind spot obstacle extraction unit 140 can correct the position of the pedestrian 21 by applying the angle θ to the position of the pedestrian 21. As a result, the position of the pedestrian 21 received from the object monitoring device 200 can be corrected to a more accurate position (FIG. 8A (3)).

When the vehicle 10 is moving toward the object monitoring device 200, the direction of the vehicle 10 detected by the object monitoring device 200 does not change substantially as the vehicle 10 moves. Therefore, since the direction of the vehicle 10 does not change even during the processing time such as the blind spot obstacle extraction unit 140 acquiring the positional relationship between the vehicle 10 and the object monitoring device 200, the position of the pedestrian 21 can be accurately corrected. .. On the other hand, when the vehicle 10 is not moving toward the object monitoring device 200 (FIG. 8A (4)), the blind spot obstacle extraction unit 140 acquires the positional relationship between the vehicle 10 and the object monitoring device 200. Within a short processing time, the direction of the vehicle 10 detected by the object monitoring device 200 changes significantly. Then, even if a slight angle θ is corrected, the effect of the correction is considered to be poor. Therefore, it is desirable that the blind spot obstacle extraction unit 140 corrects the angle θ only when the vehicle 10 is facing the object monitoring device 200.

FIG. 8B is a flowchart illustrating the detailed procedure of S704. It should be added that S704 can be omitted when the peripheral monitoring unit 210 is not a distance measuring sensor. Each step of FIG. 8B will be described below.

(FIG. 8B: step S801)
The blind spot obstacle extraction unit 140 determines whether or not the object monitoring device 200 is arranged on the extension of the traveling direction of the vehicle 10. The current position of the vehicle 10 can be obtained in S701. The traveling direction of the vehicle 10 can be determined from, for example, the steering angle of the vehicle 10. The position of the object monitoring device 200 can be described in, for example, map data mounted in advance, or can be acquired from the object monitoring device 200. If the object monitoring device 200 is arranged on the extension of the traveling direction of the vehicle 10, the process proceeds to step S802, and if not, it is considered that the correction is not possible and the present flowchart is terminated.

(FIG. 8B: Step S801: Supplement)
This step is for excluding the situation as shown in FIG. 8A (4). However, if the object monitoring device 200 is arranged at a position slightly off the extension of the traveling direction of the vehicle 10, the direction of the vehicle 10 detected by the object monitoring device 200 changes substantially as the vehicle 10 travels. It may not be. Therefore, if the angle θ formed by the line segment connecting the vehicle 10 and the object monitoring device 200 with the traveling direction of the vehicle 10 is 0 degrees or more and less than the threshold value, S802 or later may be performed. The specific threshold value may be dynamically set according to the traveling speed of the vehicle 10, for example, or a fixed value may be set in advance.

(FIG. 8B: steps S802-S804)
The blind spot obstacle extraction unit 140 acquires the position of the object monitoring device 200 (S802). The blind spot obstacle extraction unit 140 further acquires the position of the vehicle 10 from the vehicle-mounted sensor (S803) and the object monitoring device 200 (S804) mounted on the vehicle 10. As the in-vehicle sensor, for example, a position detection device such as a GPS device can be used.

(FIG. 8B: steps S805 to S806)
The blind spot obstacle extraction unit 140 is of the position of the object monitoring device 200 / the position of the vehicle 10 acquired from the vehicle-mounted sensor (first vehicle position) / the position of the vehicle 10 acquired from the object monitoring device 200 (second vehicle position). The angle θ described with reference to FIG. 8A is calculated using the three coordinates (S805). That is, the angle θ formed by the first straight line connecting the position of the object monitoring device 200 and the position of the first vehicle between the position of the object monitoring device 200 and the second straight line connecting the position of the second vehicle is calculated. Since the angle θ in S801 can be calculated in the same manner, the angle θ may be shared in S801 and S805. The blind spot obstacle extraction unit 140 corrects the position of the obstacle pair on the side other than the vehicle 10 by using the angle θ (S806).

FIG. 9 is a flowchart illustrating the operation procedure of the traveling control unit 150. Each step of FIG. 9 will be described below.

(Fig. 9: Step S901)
The travel control unit 150 acquires the positions of the obstacles detected by the vehicle 10 and the obstacles received from the object monitoring device 200. The obstacles acquired in this step include blind spot obstacles in the blind spot for the vehicle 10, objects existing in the vicinity of the vehicle 10, objects existing in the vicinity of the object monitoring device 200, and the like.

(Fig. 9: Step S902)
The travel control unit 150 calculates the time TTC [i] estimated to be required for the obstacle i to collide with the vehicle 10 (S902). The TTC [i] is calculated using, for example, the traveling direction of the vehicle 10, the speed of the vehicle 10, the distance between the obstacle i and the vehicle 10, and the like, assuming that the obstacle i is stopped. be able to. The traveling direction of the vehicle 10 can be determined from, for example, the steering angle. The speed of the vehicle 10 can be obtained from, for example, a vehicle speed sensor. When the travel control unit 150 automatically drives the vehicle 10, the speed and the traveling direction of the vehicle 10 can be acquired from the travel plan.

(Fig. 9: Step S903)
The travel control unit 150 calculates the risk level DRECI [i] indicating the possibility of the vehicle 10 colliding with the obstacle i. As a calculation method, for example, it is conceivable that the shorter the distance between the vehicle 10 and the obstacle i, the higher the DRECI [i]. The travel control unit 150 implements S901 to S903 for all obstacles i.

(FIG. 9: Steps S904 to S908)
The traveling control unit 150 identifies the obstacle k having the minimum collision prediction time TTC among the obstacles having the collision risk DRECI equal to or higher than the threshold value (S904). If the TTC [k] is equal to or less than the brake threshold value, the travel control unit 150 operates the brake via the control execution unit 160 (S905 to S906). If the TTC [k] exceeds the brake threshold value and is equal to or lower than the alarm threshold value, the travel control unit 150 outputs an alarm via the control execution unit 160 (S907 to S908). The brake threshold value is smaller than the alarm threshold value. The obstacle k is specified because it is considered that the possibility of collision with other objects can be avoided if the possibility of collision with the obstacle k can be avoided.

<Embodiment 1: Summary>
The object monitoring device 200 according to the first embodiment transmits obstacle data describing the positions of only two objects having a blind spot relationship. As a result, in the vehicle control device 100 that has received the obstacle data, the possibility that the position of the object detected by the vehicle 10 and the position of the same object detected by the object monitoring device 200 are mixed can be suppressed. Further, this makes it possible to suppress the amount of data transmitted from the object monitoring device 200 to the vehicle control device 100. In other words, since the amount of data to be processed by the vehicle control device 100 can be suppressed, the communication load and the calculation load can be suppressed.

The vehicle control device 100 according to the first embodiment determines whether or not one of the obstacle pairs described in the obstacle data is the vehicle 10, and if it is the vehicle 10, the other of the obstacle pairs. Is considered a blind spot obstacle. As a result, the positions of blind spot obstacles that cannot be recognized by the vehicle control device 100 can be acquired, and the possibility that the positions of the same objects coexist can be suppressed. Furthermore, the amount of obstacle data can be suppressed.

<Embodiment 2>
In the first embodiment, an example of detecting an object on the XY map detected by the object monitoring device 200 has been described. Since the object monitoring device 200 transmits the position only for the obstacle pair having a blind spot relationship, the amount of obstacle data can be suppressed. However, since the position of the obstacle is described by the coordinates on the XY map, if the grid of the XY map is small, a corresponding processing load is generated. Therefore, in the second embodiment of the present invention, a configuration example in which the processing load is further suppressed by roughening the grid of the coordinate system that describes the obstacle data will be described.

FIG. 10 is an example of a grid in which the coordinates around the object monitoring device 200 are described. The object monitoring device 200 sets the spatial coordinate system around the object monitoring device 200 by using a grid width larger than the spatial resolution that can be detected by the peripheral monitoring unit 210. Each rectangular area on the coarse grid is referred to as an "area" in the second embodiment. An area ID is assigned to each area.

In the second embodiment, the object monitoring device 200 uses the position of each obstacle on the rough grid as shown in FIG. 10 as obstacle data for the vehicle control device 100 instead of the coordinate system detected by the peripheral monitoring unit 210. And send. As a result, the size of obstacle data and the calculation load are suppressed.

FIG. 11 is a functional block diagram of each device according to the second embodiment. The object monitoring device 200 includes an area creation unit 260 and an area data transmission unit 270 in addition to the configuration described in the first embodiment. The vehicle control device 100 includes an area data receiving unit 170 and a position changing unit 180 in addition to the configuration described in the first embodiment.

The area creation unit 260 creates area data that describes a rough coordinate system around the object monitoring device 200 as illustrated in FIG. Area data can be described by the coordinates and ID of each area. The coordinates of the area may be described in any format as long as the vehicle control device 100 can specify the range of the area. For example, the coordinates of the four corners of the area can be described by the coordinates on GPS. Alternatively, if the size of the area is specified, the coordinates of the area can be described by the representative position of the area (for example, the upper left end). The area data transmission unit 270 transmits the area data by broadcasting. As will be described later, the obstacle data transmission unit 250 transmits the ID of the area where the obstacle exists as obstacle data.

The area data receiving unit 170 receives the area data from the object monitoring device 200. The obstacle data receiving unit 120 receives the obstacle data that describes the ID of the area where the obstacle exists. The position conversion unit 180 acquires the coordinates of the area where the obstacle exists by collating the area ID described in the obstacle data with the area data.

FIG. 12 is a schematic view showing the result of the object detection unit 220 detecting an obstacle in the second embodiment. The object detection unit 220 detects an obstacle as in the first embodiment. However, the detection result is described by the area on the area data. That is, the object detection unit 220 describes the position of the obstacle by the ID of the area where the obstacle exists. The shaded area in FIG. 12 is an area where it is determined that an obstacle exists.

FIG. 13 is a conceptual diagram showing how the blind spot relationship extraction unit 240 identifies an obstacle pair having a blind spot relationship in the second embodiment. In the first embodiment, by connecting the center of gravity of each object with a line segment, it is determined whether or not there is a shield between them. In the second embodiment, the same determination is performed by connecting the center (or the center of gravity) of the area where the obstacle exists with a line segment instead of the center of gravity of the object. The coordinates of the area are specified by the position conversion unit 180. Subsequent processing is the same as in the first embodiment.

FIG. 14 is a conceptual diagram showing how area data and obstacle data are transmitted and received between the object monitoring device 200 and the vehicle control device 100. Since the coordinate system based on the coarse grid described with reference to FIG. 10 is created by the object monitoring device 200, it is necessary to notify the vehicle control device 100 of the coordinates and ID of each area. The object monitoring device 200 transmits the area data, and the vehicle control device 100 receives the area data to recognize the coordinates and ID of the area set by the object monitoring device 200. Since it is sufficient that the obstacle data can identify the area where the obstacle exists, it can be described by the two area IDs where the two obstacles exist.

The object monitoring device 200 may transmit the area data and the obstacle data at the same time, or may transmit, for example, alternately. However, once the vehicle control device 100 receives the area data, the area data can be reused until the object monitoring device 200 changes the area configuration. Therefore, the transmission frequency of the area data may be lower than the transmission frequency of the obstacle data.

<Embodiment 2: Summary>
The object monitoring device 200 according to the second embodiment transmits area data in which the peripheral coordinates of the object monitoring device 200 are divided into a plurality of areas and described, and also uses the ID of the area where the obstacle exists as the obstacle data. Send. As a result, the size of obstacle data can be suppressed, and the communication load and the calculation load can be suppressed.

<Embodiment 3>
In the first and second embodiments, an example of transmitting and receiving obstacle data between the vehicle control device 100 and the object monitoring device 200 has been described. Instead of or in addition to this, the vehicle control device 100 has the same configuration as the object monitoring device 200, and obstacle data can be transmitted and received between the vehicle control devices 100 mounted on each vehicle. In the third embodiment of the present invention, a configuration example thereof will be described.

FIG. 15 is a functional block diagram of the vehicle control device 100 according to the third embodiment. The vehicle control device 100 in the third embodiment includes each functional unit included in the object monitoring device 200 in addition to the configurations described in the first and second embodiments. Here, the same configuration example as in the first embodiment is shown. However, the peripheral monitoring unit 110 and the object detection unit 130 are replaced by the peripheral monitoring unit 210 and the object detection unit 220, respectively.

The obstacle data transmission unit 250 transmits obstacle data by broadcasting, and the other vehicle control device 100 receives this. The obstacle data receiving unit 120 receives obstacle data from another vehicle control device 100. Other configurations are the same as those of the first and second embodiments. By transmitting and receiving obstacle data between the vehicle control devices 100, it is possible to recognize an object that is a blind spot for the vehicle control device 100 even in a place where communication with the object monitoring device 200 is not possible. Further, the size of obstacle data transmitted / received between the vehicle control devices 100 can be suppressed.

<About a modified example of the present invention>
The present invention is not limited to the above embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration.

In the second embodiment, the orthogonal coordinate system is illustrated as the coordinate system described by the area data, but the configuration of the area is not limited to this. Area data can also be described by other coordinate systems such as a rotating coordinate system.

In the above embodiment, each functional unit included in the vehicle control device 100 and the object monitoring device 200 can be configured by hardware such as a circuit device that implements the function, or software that implements the function is used as an arithmetic unit. It can also be configured by executing (such as Central Processing Unit).

1: Vehicle control system 10: Vehicle 21: Pedestrian 22: Pedestrian 30: Obstruction 100: Vehicle control device 110: Peripheral monitoring unit 120: Obstacle data reception unit 130: Object detection unit 140: Blind spot obstacle extraction unit 150 : Travel control unit 160: Control execution unit 170: Area data reception unit 180: Position conversion unit 200: Object monitoring device 210: Peripheral monitoring unit 220: Object detection unit 230: Topography acquisition unit 240: Blind spot relationship extraction unit 250: Obstacle Data transmission unit 260: Area creation unit 270: Area data transmission unit

Claims (15)

A vehicle control device that controls the operation of a vehicle.
Peripheral monitoring unit that detects the position of objects around the vehicle existing around the vehicle,
Obstacle to receive obstacle data describing a pair of the first object position of the first object detected by the object monitoring device installed outside the vehicle and the second object position of the second object detected by the object monitoring device. Object data receiver,
It is determined whether or not the first object described in the obstacle data is the vehicle, and when the first object is the vehicle, the peripheral monitoring unit cannot detect the first object as a blind spot obstacle. Blind spot obstacle extraction unit, which acquires the position of the second object of two objects,
A vehicle control device characterized by being equipped with. The vehicle control device further includes a travel control unit that creates a travel plan for the vehicle.
The travel control unit uses the object position, the position of the blind spot obstacle, and the travel plan to determine whether or not the vehicle collides with the object around the vehicle and whether the vehicle collides with the blind spot obstacle. Perform a collision judgment to determine whether or not
The traveling control unit calculates a collision prediction time required for the vehicle to collide with an object around the vehicle or the blind spot obstacle with the vehicle.
When the collision prediction time is equal to or less than the first threshold value, the traveling control unit reduces the speed of the vehicle by operating the brake of the vehicle.
The vehicle control device according to claim 1, wherein the traveling control unit outputs an alert indicating that when the collision prediction time is larger than the first threshold value and is equal to or less than the second threshold value. The traveling control unit calculates a collision risk indicating the possibility of the vehicle colliding with the vehicle peripheral object or the blind spot obstacle.
The traveling control unit identifies the dangerous object having the highest collision risk among the vehicle peripheral object and the blind spot obstacle.
2. The traveling control unit controls the vehicle so that the possibility of the dangerous object colliding with the vehicle is reduced by calculating the collision prediction time for the dangerous object. The vehicle control device described. The vehicle control device further includes an area data receiving unit that receives area data described by dividing the peripheral coordinates of the vehicle into a plurality of areas.
The obstacle data receiving unit receives, as the obstacle data, data describing a pair of a first ID of the area where the first object exists and a second ID of the area where the second object exists. The vehicle control device according to claim 1. The blind spot obstacle extraction unit acquires the position of the object monitoring device and the first vehicle position of the vehicle from the object monitoring device.
The blind spot obstacle extraction unit acquires the position of the second vehicle of the vehicle from the position detection device included in the vehicle.
The blind spot obstacle extraction unit is formed by a first straight line connecting the position of the object monitoring device and the position of the first vehicle between the position of the object monitoring device and the second straight line connecting the position of the second vehicle. The vehicle control device according to claim 1, wherein when the angle is greater than 0 and less than or equal to a predetermined angle, the position of the second object is corrected using the angle. The vehicle control device further
It is determined whether or not there is a shield between the third object detected by the peripheral monitoring unit and the fourth object detected by the peripheral monitoring unit, and if so, the third object and the third object are present. Blind spot relationship extraction unit, which determines that four objects have a blind spot relationship in which they are blind spots to each other.
An obstacle data transmission unit that transmits obstacle data that describes the positions of the third object and the fourth object.
With
When the third object and the fourth object have the blind spot relationship, the obstacle data transmission unit transmits the obstacle data.
The vehicle control device according to claim 1, wherein the obstacle data transmission unit does not transmit the obstacle data when the third object and the fourth object do not have the blind spot relationship. An object monitoring device installed outside the vehicle to monitor objects existing around the vehicle.
Peripheral monitoring unit that detects the position of peripheral objects existing around the object monitoring device,
It is determined whether or not there is a shield between the first object detected by the peripheral monitoring unit and the second object detected by the peripheral monitoring unit, and if so, the first object and the first object are present. A blind spot relationship extraction unit that determines that two objects have a blind spot relationship that is a blind spot to each other.
An obstacle data transmission unit that transmits obstacle data that describes the positions of the first object and the second object.
With
When the first object and the second object have the blind spot relationship, the obstacle data transmission unit transmits the obstacle data.
The obstacle data transmission unit is an object monitoring device that does not transmit the obstacle data when the first object and the second object do not have the blind spot relationship. The peripheral monitoring unit is configured to detect the boundary position of the peripheral object.
The object monitoring device further includes an object detection unit that identifies a range of the peripheral object using the boundary position detected by the peripheral monitoring unit.
When the height of the boundary position from the ground is within a predetermined range, the object detection unit identifies the boundary position as the boundary of the peripheral object.
The object monitoring device according to claim 7, wherein the object detection unit does not treat the boundary position as a boundary of the peripheral objects when the height of the boundary position from the ground is outside the predetermined range. .. The object monitoring device further includes an object detection unit that identifies a peripheral object whose position does not change as a fixed object and creates a background map that describes the coordinates of the fixed object.
The object detection unit creates an average map by averaging an object map that describes the positions of the surrounding objects, and the average map until the dispersion of the object positions described by the average map becomes less than a predetermined threshold. 7. The object monitoring device according to claim 7, wherein the background map is created by repeatedly creating the above. The object detection unit is characterized in that it identifies a moving object or a temporarily placed object from the peripheral objects by obtaining a difference between the position of the peripheral object and the background map. 9. The object monitoring device according to claim 9. The object monitoring device further includes a terrain acquisition unit that identifies a fixed object having a height equal to or higher than a predetermined height as a candidate for the shield.
The object monitoring device according to claim 9, wherein the blind spot relationship extraction unit determines the presence or absence of the blind spot relationship by using the candidate of the shield specified by the terrain acquisition unit. The object monitoring device further includes an object detection unit that identifies a peripheral object whose position does not change as a fixed object and creates a background map that describes the coordinates of the fixed object.
The object detection unit creates the background map as a two-dimensional map by specifying the plane coordinates of the fixed object.
The object monitoring device according to claim 11, wherein the terrain acquisition unit identifies a candidate for the shield by collating a height map describing the height of the fixed object with the background map. The object monitoring device further includes a terrain acquisition unit that identifies a fixed object having a height equal to or higher than a predetermined height as a candidate for the shield.
The blind spot relationship extraction unit obtains a straight line connecting the first object and the second object on the background map, and whether or not the candidate for the shield specified by the terrain acquisition unit exists on the straight line. 9. The object monitoring device according to claim 9, wherein it is determined whether or not the first object and the second object have the blind spot relationship. The object monitoring device further includes an area data creation unit that creates area data in which the peripheral coordinates of the object monitoring device are divided into a plurality of areas and described.
The object monitoring device further includes an area data transmission unit for transmitting the area data.
The obstacle data transmission unit transmits, as the obstacle data, data describing a pair of a first ID of the area where the first object exists and a second ID of the area where the second object exists. The object monitoring device according to claim 7, wherein the object monitoring device is characterized. A vehicle control system that controls the operation of a vehicle.
A vehicle control device that controls the operation of the vehicle,
An object monitoring device installed outside the vehicle to monitor objects existing around the vehicle,
With
The vehicle control device
Peripheral monitoring unit that detects the position of objects around the vehicle existing around the vehicle,
Obstacle to receive obstacle data describing a pair of the first object position of the first object detected by the object monitoring device installed outside the vehicle and the second object position of the second object detected by the object monitoring device. Object data receiver,
It is determined whether or not the first object described in the obstacle data is the vehicle, and when the first object is the vehicle, the peripheral monitoring unit cannot detect the first object as a blind spot obstacle. Blind spot obstacle extraction unit, which acquires the position of the second object of two objects,
With
The object monitoring device is
Peripheral monitoring unit that detects the position of peripheral objects existing around the object monitoring device,
It is determined whether or not there is a shield between the third object detected by the peripheral monitoring unit and the fourth object detected by the peripheral monitoring unit, and if so, the third object and the third object are present. Blind spot relationship extraction unit, which determines that four objects have a blind spot relationship in which they are blind spots to each other.
An obstacle data transmission unit that transmits obstacle data that describes the positions of the third object and the fourth object.
With
When the third object and the fourth object have the blind spot relationship, the obstacle data transmission unit transmits the obstacle data.
The vehicle control system is characterized in that the obstacle data transmission unit does not transmit the obstacle data when the third object and the fourth object do not have the blind spot relationship.

Images