Vehicle controller

Abstract

To provide a vehicle controller capable of easily determining whether a target object is split or not when a point cloud belonging to the target object is recognized as being divided into a plurality of point clouds.SOLUTION: The vehicle controller according to the present invention includes: recognition means for recognizing a target object around a vehicle as point cloud data; setting means for setting a box so that the box encloses the recognized point cloud data; and determination means for determining whether a target object indicated by one of a plurality of divided boxes is split when the box is divided into the plurality of divided boxes.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a vehicle control device. [Background technology]

[0002] In recent years, research has been progressing on autonomous driving, which allows a vehicle to travel without user operation. For autonomous driving control, for example, a technology has been disclosed relating to a vehicle control device that recognizes targets as point cloud data using a light detection and ranging (LiDAR) mounted on the vehicle and identifies targets present around the vehicle from the recognized point cloud data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-32069 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when identifying a target from point cloud data, for example, a point cloud belonging to the same target may be recognized as being split into multiple point clouds due to a change in the intensity of light reflected from the target surface, etc. In such cases, with conventional techniques, it has been difficult to determine whether the target identified by the point cloud is actually split.

[0005] The object of the present invention has been made in consideration of the above-mentioned problems, and is to provide a vehicle control device that can easily determine whether a target has been split when a point cloud belonging to the target is recognized as being split into multiple point clouds. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the vehicle control device of the present invention comprises a recognition means for recognizing targets around the vehicle as point cloud data, a setting means for setting a box to surround the recognized point cloud data, and a determination means for determining whether or not the target indicated by one of the multiple split boxes has split when the box is split into multiple split boxes. [Effects of the Invention]

[0007] According to the present invention, when a point cloud belonging to a target is recognized as being split into a plurality of point clouds, it is possible to easily determine whether the target is split or not. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing an example of a system configuration of a vehicle equipped with a vehicle control device according to an embodiment. [Figure 2] FIG. 2 is a block diagram illustrating an example of the functional configuration of the autonomous driving ECU of the vehicle according to the embodiment. [Figure 3] FIG. 3 is a diagram showing an example in which a rectangular parallelepiped shape fitted to a target object is divided into multiple parts. [Figure 4] FIG. 4 is a block diagram illustrating an example of a functional configuration of an object recognition unit according to the embodiment. [Figure 5] FIG. 5 is a flowchart showing the procedure of determination by the determining means. [Figure 6] FIG. 6 is a diagram illustrating the relationship between a box before division and a plurality of division boxes. [Figure 7] FIG. 7 is a diagram illustrating how a box before division is superimposed on multiple division boxes. [Figure 8] FIG. 8 is a diagram illustrating the overlapping portion between the box before division and the division box. [Figure 9] FIG. 9 is a flowchart showing the procedure for repeatedly making determinations based on each division box. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The configuration of the embodiment described below and the actions and effects brought about by the configuration are merely examples, and the present invention is not limited to the following description.

[0010] 1 is a block diagram showing an example of a system configuration of a vehicle 1 equipped with a vehicle control device according to an embodiment. The vehicle 1 is equipped with an automatic driving function and is capable of traveling by automatic driving without the need for driving operations by a user (driver). Note that automatic driving includes semi-automatic driving in which some of the operations for traveling of the vehicle 1 are automated (requiring partial driving operations by the user).

[0011] A plurality of ECUs (Electronic Control Units) are mounted on the vehicle 1 to control various parts. Each ECU has a microcontroller unit (microcomputer), and the microcomputer has built-in, for example, a CPU (Central Processing Unit), a non-volatile memory such as a flash memory, and a volatile memory such as a DRAM (Dynamic Random Access Memory).

[0012] The multiple ECUs include a drive ECU 11, a steering ECU 12, a brake ECU 13, a meter ECU 14, and a body ECU 15. The drive ECU 11, the steering ECU 12, the brake ECU 13, the meter ECU 14, and the body ECU 15 are connected to each other so as to be able to communicate using a CAN (Controller Area Network) communication protocol, that is, to perform CAN communication.

[0013] The drive ECU 11 is a control unit that controls a drive unit 21 of the vehicle 1. The drive unit 21 may be configured to include an engine as a drive source, a motor as a drive source, or both an engine and a motor as drive sources. The drive unit 21 includes a transmission that changes the speed of the drive force from the drive source and outputs it as necessary.

[0014] The steering ECU 12 is a control unit that controls a steering device 22 of the vehicle 1. The steering device 22 is, for example, an electric power steering device that applies torque from an electric motor to a steering mechanism. The steering mechanism includes, for example, a rack-and-pinion steering gear, and is configured so that when a rack shaft moves in the vehicle width direction due to the torque of the electric motor, the left and right steered wheels are turned left and right in accordance with the movement of the rack shaft.

[0015] The brake ECU 13 is a control unit that controls the braking device 23 of the vehicle 1. The braking device 23 may be of a hydraulic or electric type. The hydraulic braking device 23 is equipped with a brake actuator, and the function of this brake actuator distributes hydraulic pressure to the wheel cylinders of the brakes provided on each wheel, and the hydraulic pressure drives the brakes from each brake to the drive wheels. A braking force is applied to the wheels including the

[0016] The meter ECU 14 is a control unit that controls each part of a meter panel (not shown) of the vehicle 1. The meter panel is provided with indicators such as a liquid crystal display for displaying various information, as well as instruments that display vehicle speed and engine RPM. An emergency stop switch 24 that is operated to issue an emergency stop command for the autonomous driving is also connected to the meter ECU 14.

[0017] The body ECU 15 is a control unit that controls various parts that need to operate even when the ignition switch of the vehicle 1 is off, such as the left and right turn signals and door lock motors.

[0018] The plurality of ECUs also include an automatic driving ECU 31, a lidar ECU 32, and a monocular camera ECU 33 as control units for the automatic driving function.

[0019] The autonomous driving ECU 31 is a control center for autonomous driving control. The autonomous driving ECU 31 is an example of a vehicle control device. The autonomous driving ECU 31 is connected to the drive ECU 11, the steering ECU 12, the brake ECU 13, the meter ECU 14, and the body ECU 15 so as to be able to communicate via CAN.

[0020] An omnidirectional LiDAR (Light Detection and Ranging) 34 is connected to the autonomous driving ECU 31 via, for example, an Ethernet (registered trademark) communication cable. The omnidirectional LiDAR 34 is an example of a "distance measurement sensor" capable of acquiring distance measurement information indicating the distance from the vehicle 1 to objects present around the vehicle 1. The omnidirectional LiDAR 34 emits laser light in all directions of 360°, receives reflected light from objects present within a search range with an optical sensor, and outputs a detection signal corresponding to the reflected light as distance measurement information. The distance measurement information may take the form of, for example, point cloud information indicating the distance from the vehicle 1 to the object at each position (voxel) in three-dimensional space. A detection signal from the omnidirectional LiDAR 34 is input to the autonomous driving ECU 31. The omnidirectional LiDAR 34 is mounted at a relatively high position on the vehicle 1 and is capable of receiving reflected light from the depth direction of targets such as other vehicles.

[0021] Furthermore, a GPS receiver 35 is connected to the autonomous driving ECU 31 via, for example, a USB (Universal Serial Bus) standard communication cable. The GPS receiver 35 is a receiver that receives positioning signals from GPS (Global Positioning System) satellites. The GPS receiver 35 is an example of a "positioning sensor" that can acquire positioning information indicating a position on Earth (e.g., latitude, longitude, altitude, etc.). The positioning signals received by the GPS receiver 35 are input from the GPS receiver 35 to the autonomous driving ECU 31 as positioning information.

[0022] The LIDAR ECU 32 is communicatively connected to the autonomous driving ECU 31 via, for example, an Ethernet standard communication cable. Six LIDARs 36 are connected to the LIDAR ECU 32. Each LIDAR 36 irradiates a search range with laser light, receives reflected light from objects present within the search range with an optical sensor, and outputs a detection signal corresponding to the reflected light. The LIDARs 36 are, for example, disposed at the left end, center, and right end of the front bumper and the left end, center, and right end of the rear bumper of the vehicle 1. The LIDARs 36 are an example of distance measuring sensors. The LIDAR ECU 32 receives detection signals output from each LIDAR 36. The LIDAR ECU 32 processes the detection signals output from each LIDAR 36 and transmits data obtained by the processing to the autonomous driving ECU 31.

[0023] The monocular camera ECU 33 is communicably connected to the autonomous driving ECU 31 via, for example, a USB standard communication cable. A monocular camera 37 is connected to the monocular camera ECU 33. The monocular camera 37 is a camera that can continuously capture still images of the search range ahead of the vehicle 1 at a predetermined frame rate. Image signals of the still images continuously output from the monocular camera 37 are input to the monocular camera ECU 33. The monocular camera ECU 33 processes the image signals input from the monocular camera 37 and transmits image data obtained by this processing to the autonomous driving ECU 31.

[0024] 2 is a block diagram showing an example of the functional configuration of the autonomous driving ECU 31 of the vehicle 1 according to this embodiment. The autonomous driving ECU 31 of this embodiment includes an object recognition unit 41, a self-position estimation unit 42, a surrounding information integration unit 43, a route planning unit 44, and a vehicle control unit 45. These functional units 41 to 45 are configured by cooperation between hardware and software (programs, etc.) that configure the vehicle 1 as shown in FIG. 1. Furthermore, at least one of these functional units 41 to 45 may be configured by dedicated hardware (circuits, etc.). However, the functional configuration of the autonomous driving ECU 31 is not limited to this.

[0025] The object recognition unit 41 recognizes targets such as other vehicles and pedestrians around the vehicle 1 from information on the distance to the targets (vehicles, pedestrians, buildings, curbs, and other obstacles) obtained from the detection signal of the omnidirectional lidar 34 and from images captured by the monocular camera 37.

[0026] For example, the object recognition unit 41 recognizes targets by performing the following processes. Specifically, the object recognition unit 41 performs a ground removal process to remove point clouds due to reflection from the ground from the point cloud data of the omnidirectional LIDAR 34. The object recognition unit 41 also performs a clustering process to group the point clouds into points that are close to each other. Furthermore, the object recognition unit 41 performs a boxing process to fit the grouped point clouds into a rectangular parallelepiped shape in order to recognize them as target objects that are present around the vehicle and are to be detected.

[0027] The object recognition unit 41 then performs tracking processing to calculate the relative speed of the target object by monitoring the boxed rectangular parallelepiped shape. This allows the object recognition unit 41 to recognize the target. Note that the boxed information obtained by fitting the grouped point clouds to the rectangular parallelepiped shape is also referred to as surrounding point cloud information indicating the surrounding point cloud. The object recognition unit 41 also assigns an ID (Identification) to the rectangular parallelepiped shape and monitors the target. For example, a rectangular parallelepiped shape assigned the nth ID is monitored as a rectangular parallelepiped shape with ID=n.

[0028] The self-position estimation unit 42 matches the point cloud data acquired from the detection signals of the omnidirectional LIDAR 34 with high-precision map data (point cloud data) 46, which is data of a high-precision map, to estimate the position (self-position) of the vehicle 1. The high-precision map is a high-precision three-dimensional map, and the high-precision map data 46 includes information such as road width and gradient, and information on features such as dividing lines, shoulder lines, intersections, railroad crossings, stop lines, crosswalks, and signs.

[0029] The high-precision map data 46 may be stored in a non-volatile memory built into the microcomputer of the autonomous driving ECU 31, or may be stored in an HDD (Hard Disk Drive) or the like connected to the autonomous driving ECU 31. The self-position estimation unit 42 integrates the self-position estimated by matching the point cloud data with the self-position based on the positioning signal received by the GPS receiver 35, thereby improving the estimation accuracy of the self-position.

[0030] The surrounding information integrating unit 43 receives as input the target recognition results from the object recognition unit 41, the self-position estimation results from the self-position estimation unit 42, and data obtained by processing detection signals output from each LIDAR 36 in the LIDAR ECU 32 (see FIG. 1). High-precision map data 46 is also input to the surrounding information integrating unit 43. The surrounding information integrating unit 43 creates surrounding information integrated map data in which targets such as the vehicle 1, vehicles other than the vehicle 1, and pedestrians are arranged on a high-precision map. The surrounding information integrating unit 43 then outputs the map information and object recognition information to an HMI device 47 (Human Machine Interface) such as a display provided inside the vehicle 1.

[0031] The route planning unit 44 receives the peripheral information integrated map data from the peripheral information integrating unit 43. The route planning unit 44 plans a travel route to the destination of the vehicle 1 from the input peripheral information integrated map data. The travel route includes the path of the vehicle 1 and the target vehicle speed at each point on the path. The route planning unit 44 outputs route data of the planned travel route to the HMI device 47.

[0032] The vehicle control unit 45 receives route data from the route planning unit 44. Based on the route data, the vehicle control unit 45 outputs commands to ECUs that control the operation of each part of the vehicle 1, such as the drive ECU 11, the steering ECU 12, and the brake ECU 13, so that the vehicle 1 travels automatically along the travel route.

[0033] For example, after the destination of vehicle 1 is input on HMI device 47, an autonomous driving start button displayed on HMI device 47 is pressed, whereby an instruction to start autonomous driving is input from HMI device 47 to autonomous driving ECU 31. When the instruction to start autonomous driving is input to autonomous driving ECU 31, a travel route to the destination of vehicle 1 is planned by route planning unit 44. The travel route is re-planned at predetermined intervals while vehicle 1 is traveling autonomously.

[0034] Then, vehicle 1 travels along the most recently planned travel route at the target vehicle speed for each point on the route. The autonomous driving ends, for example, when vehicle 1 arrives at the destination or when emergency stop switch 24 is pressed and an instruction to stop the autonomous driving is input from meter ECU 14 to autonomous driving ECU 31.

[0035] In some cases, the object recognition unit 41 may observe that a monitored target appears to split into multiple pieces at a certain point. This occurs because a single point cloud belonging to the target is grouped into multiple point clouds due to, for example, changes in the intensity of reflected light from the target. FIG. 3 is a diagram illustrating an example in which a rectangular parallelepiped shape fitted to a target is split into multiple pieces. FIG. 3 is a plan view of a vehicle, which is the target, seen from directly above, and the rectangular parallelepiped shape fitted to the vehicle is indicated by a rectangular frame (box). At time t1, one box (ID=1) is assigned to the vehicle, but at a later time t2, two boxes (ID=1, ID=2) are assigned to the same vehicle. Because it appears as if one box has split into two, these will hereinafter be referred to as "split boxes." In this example, the split box behind the vehicle is linked to the original box and assigned the same ID (ID=1), while the split box in front of the vehicle is assigned a different ID (ID=2). In the prior art, in such a case, it was difficult to determine whether the target object had split at time t2. Furthermore, if a target object that was not actually split was determined to have split, the center of gravity of the target would be shifted, causing an error in the relative velocity calculated by the tracking process, which could lead to an erroneous stop or deceleration of the vehicle.

[0036] In this embodiment, the above problem is solved by determining whether the target object is split or not using the box overlap ratio, as will be described in detail below.

[0037] 4 is a block diagram showing an example of the functional configuration of the object recognition unit 41 according to this embodiment. The object recognition unit 41 includes a recognition unit 411, a setting unit 412, and a determination unit 413. Note that the functions included in the object recognition unit 41 are not limited to these. Furthermore, in this embodiment, these functional units 411 to 413 are described as being included in the object recognition unit 41, but other functional units of the autonomous driving ECU 31 may include the recognition unit 411, the setting unit 412, and the determination unit 413.

[0038] The recognition means 411 recognizes targets around the vehicle as point cloud data. The point cloud data is one of the plurality of point clouds grouped as described above.

[0039] The setting means 412 fits a rectangular parallelepiped shape to surround the point cloud data recognized by the recognition means 411, and sets a box by projecting this onto a horizontal plane. An ID such as that shown in FIG. 3 is associated with the box set for each point cloud data. The shape of the box is not limited to a rectangle. For example, if the bottom surface of the rectangular parallelepiped is inclined with respect to the horizontal plane, the shape of the box will be polygonal. Furthermore, if the point cloud data is fitted to a shape other than a rectangular parallelepiped, the shape of the box will be a shape other than a rectangle.

[0040] At time t2, when the box (ID=1) set at time t1 splits into multiple split boxes (ID=1, ID=2), the determination means 413 determines whether the targets indicated by these multiple boxes have actually split using the method described below.

[0041] 5 is a flowchart showing the procedure of judgment in the judgment means 413. First, the judgment means 413 identifies a reference division box (step S10) and overlays the pre-division box on multiple division boxes (step S11). The reference division box is a box linked to the pre-division box, and both boxes have the same ID. Furthermore, as will be described later, the pre-division box is overlaid on multiple division boxes, with a vertex of the reference division box serving as a base point. Here, the reference division box is an example of a first division box.

[0042] The box before splitting is the box that was set before time t2 when the split box was set. The box before splitting may be the box that the setting means 412 set just before time t2, or may be a box that was set at an earlier time. For example, it may be the box with the largest area among the boxes that the setting means 412 set in the predetermined number of settings before t2. Here, the predetermined number is, for example, around five times, and may be determined in advance through experiments or the like, or may be dynamically changed during vehicle control operation.

[0043] FIG. 6 is a diagram illustrating the relationship between a box before division and multiple division boxes. FIG. 6(a) shows that a box (ID=1) set at time t1 has divided into multiple division boxes at time t2. FIG. 6(b) is an example of a box before division (a box set before time t2) that is overlaid on multiple division boxes. FIG. 6(c) shows a case where the lower division box of two division boxes at time t2 is linked to the box before division, and FIG. 6(d) shows a case where the upper division box is linked to the box before division.

[0044] FIG. 7 is a diagram illustrating how a pre-split box is overlaid on multiple split boxes. As in FIG. 6(c), when the lower split box is linked to the pre-split box, the lower split box is identified as the reference split box. In this case, the determination means 413 overlays multiple split boxes by using the lower left vertex of the split box (ID=1) as the base point and aligning the lower left vertex of the pre-split box as the base point, as in FIG. 7(a). Note that the lower right vertex of the split box (ID=1) may also be used as the base point, and in that case, the determination means 413 overlays multiple split boxes by using the lower right vertex of the pre-split box as the base point, as in FIG. 7(b).

[0045] As shown in Figure 6(d), if the upper division box is linked to the box before division, the upper division box is identified as the reference division box. In this case, the determination means 413 uses the upper left vertex of the division box (ID = 1) as the base point, and overlaps multiple division boxes with the upper left vertex of the box before division as the base point, as shown in Figure 7(c). Note that the upper right vertex of the division box (ID = 1) may also be used as the base point, and in that case, the determination means 413 overlaps multiple division boxes with the upper right vertex of the box before division as the base point, as shown in Figure 7(d).

[0046] In addition, if the shape of the box is other than rectangular, for example, the base point of the division box (ID=1) can be set and the vertex of the box before division can be aligned with the base point so that the overlapping area between the box before division and the division box (ID=1) is maximized.

[0047] Next, the determination means 413 calculates the overlap ratio between the box before division and one of the multiple division boxes (step S12). Specifically, the overlap ratio is determined by dividing the area of ​​the overlap between the box before division and a division box (ID=2) that is not linked to the box before division by the area of ​​the division box (ID=2). Figure 8 is a diagram explaining the overlap between the box before division and the division box. Each of Figures 8(a) to (d) shows the overlap between the box before division and the division box (ID=2) with diagonal lines for each of the cases of Figures 7(a) to (d). Here, the division box that is not linked to the box before division is an example of a second division box.

[0048] If the overlap ratio calculated in step S12 is equal to or greater than a predetermined value (R) (step S13: Yes), the determination means 413 determines that the target indicated by the division box (ID=2) for which the overlap ratio was calculated is not a split target (step S14). That is, the determination means 413 determines that the target indicated by the division box (ID=2) for which the overlap ratio was calculated is the same target as the target indicated by the division box (ID=1), and is linked to the box before splitting (ID=1). On the other hand, if the overlap ratio is smaller than the predetermined value (R) (step S13: No), the determination means 413 determines that the target indicated by the division box (ID=2) for which the overlap ratio was calculated is a split target (step S15). That is, the determination means 413 determines that the target indicated by the division box (ID=2) for which the overlap ratio was calculated is a different target from the target indicated by the division box (ID=1). Here, the value R is a predetermined ratio, for example, around 50%, and may be determined in advance through experiments or the like, or may be dynamically changed during the operation of vehicle control.

[0049] If the determination means 413 determines that the target is not a split object (step S14), the object recognition unit 41 applies the point clouds belonging to the set multiple split boxes (ID=1, ID=2) as one group to a rectangular parallelepiped shape and sets one box (ID=1). Then, it continues monitoring the box (ID=1). On the other hand, if the determination means 413 determines that the target is a split object (step S15), the object recognition unit 41 monitors the multiple targets indicated by the set multiple split boxes (ID=1, ID=2) as separate targets.

[0050] In the above example, the overlap rate was calculated after overlapping the pre-split box with the split box, but it is not necessary to actually overlap the boxes. The overlap rate can be calculated by using the shape of each box (side length, vertex position, etc.) to calculate the shape of the overlapping area when the boxes are overlapped.

[0051] Furthermore, in the examples of Figures 6(c) and (d), the case where the ID of the split box can be determined has been described. However, there are also cases where the ID of the split box cannot be determined (when it is unclear which split box the box before splitting is linked to). In such cases, each split box is selected as a reference split box, and the above-mentioned determination is repeated to determine whether or not the target is a split object. For example, when a box splits as shown in Figure 6(a), first, the lower split box is selected as the reference split box as shown in Figure 7(a) or (b). If, as a result of the above determination, it is determined that the target indicated by the upper split box is not a split object, the determination process is terminated. Here, the lower split box is an example of a first split box, and the upper split box is an example of a second split box.

[0052] On the other hand, if it is determined that the target is a split target, then the upper split box is selected as the reference split box, as shown in Figure 7(c) or (d), and the above-mentioned determination is made to determine whether the target indicated by the lower split box is a split target. Here, the upper split box is an example of the third split box, and the lower split box is an example of the fourth split box.

[0053] Figure 9 is a flowchart showing the procedure for repeatedly making judgments based on each division box. The difference from Figure 5 is that if the overlap rate is smaller than a predetermined value (R) (step S23: No), it is determined whether all division boxes have been selected as division boxes to be used as the reference (step S25), and the processing of steps S20 to S25 is repeated for the remaining division boxes. The rest of the process is the same as Figure 5, so detailed explanations will be omitted.

[0054] 6 to 8 show examples where there are two division boxes, but there may be three or more division boxes. For example, the process of FIG. 5 can be applied when there are three or more division boxes by repeating steps S12 to S15 for each division box other than the reference division box. Furthermore, the process of FIG. 9 can be applied when there are three or more division boxes by repeating steps S22 to S25 for each division box other than the division box selected in step S20.

[0055] As described above, according to this embodiment, if the overlap between the pre-split box and the split box is equal to or greater than a predetermined percentage, the target indicated by the split box is determined to be unsplit. This makes it easy to determine whether the target has split when a point cloud belonging to the target is split into multiple point clouds and recognized, preventing erroneous determination that the point cloud or box has split. This prevents erroneous recognition due to a sudden change in the center of gravity of the monitored box, enabling stable target tracking processing and more appropriate vehicle control. Furthermore, since the determination process of this embodiment does not require the target's speed or movement direction, it can be applied to boxes set for stationary targets, and the simple process reduces the amount of determination processing required.

[0056] In addition, since the overlap ratio when the division box linked to the box before the division is used as a reference is used to make the determination, it is possible to appropriately determine whether or not a division has occurred. Furthermore, since the determination can be made by repeating the process using each division box as a reference, it is possible to appropriately determine whether or not a division has occurred even if the division box linked to the box before the division is unknown.

[0057] Furthermore, if the box before division is set to the box with the largest area among the boxes set in a predetermined number of settings before division, the area of ​​the part that overlaps with the division box can be made larger, thereby increasing the possibility of determining that there has been no division, and enabling more stable tracking processing.

[0058] A program that causes a computer (e.g., the autonomous driving ECU 31, etc.) to execute processes for realizing various functions in the control device of the vehicle 1 as described above can be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD (Compact Disc)-ROM (Read Only Memory), a flexible disk (FD), a CD-R (Recordable), or a DVD (Digital Versatile Disk). The program may also be provided or distributed via a network such as the Internet. The program may also be provided by being pre-installed in a ROM or the like.

[0059] Although the embodiments of the present invention have been described above, the above-described embodiments are presented as examples and are not intended to limit the scope of the present invention. This novel embodiment can be embodied in various other forms. Furthermore, various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. Furthermore, this embodiment is included within the scope and spirit of the invention, and is also included in the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0060] 1 vehicle 11 Drive ECU 12 Steering ECU 13 Brake ECU 14 Meter ECU 15 Body ECU 24 Emergency stop switch 31 Autonomous Driving ECU 32 Rider ECU 33 Monocular camera ECU 34 Omnidirectional Lidar 35 GPS receiver 36 Rider 37 Monocular Camera 41 Object recognition section 42 Self-position estimation part 43 Peripheral Information Integration Department 44 Route Planning Section 45 Vehicle control unit 411 Recognition means 412 Setting Method 413 Judgment means

Claims

1. recognition means for recognizing targets around the vehicle as point cloud data; a setting means for setting a box so as to surround the recognized point cloud data; a determination means for determining whether a target indicated by one of the plurality of division boxes is divided when the box is divided into a plurality of division boxes; Equipped with The vehicle control device has a determination means that, when the box before splitting is superimposed on the multiple split boxes, if the overlapping portion between the box before splitting and one of the split boxes is equal to or greater than a predetermined percentage, determines that the target indicated by the one of the split boxes is not split.

2. The determination means When the box before division is overlapped on the plurality of division boxes with the vertex of a first division box as a base point, If an overlapping portion between the box before division and a second division box other than the first division box among the plurality of division boxes is equal to or greater than the predetermined ratio, it is determined that the target indicated by the second division box is not a divided target. The vehicle control device according to claim 1 .

3. The determination means When the box before division is overlapped on the plurality of division boxes with the vertex of a first division box as a base point, If an overlapping portion between the box before division and a second division box other than the first division box among the plurality of division boxes is equal to or greater than the predetermined ratio, it is determined that the target indicated by the second division box is not a divided target; When the box before division is overlapped on the plurality of division boxes, using a vertex of a third division box other than the first division box as a base point, If the overlapping portion between the box before division and a fourth division box other than the third division box among the plurality of division boxes is equal to or greater than the predetermined ratio, it is determined that the target indicated by the fourth division box is not a divided target. The vehicle control device according to claim 1 .

4. the determining means uses the box having the largest area among the boxes set by the setting means in a predetermined number of settings before the division as the box before the division; The vehicle control device according to any one of claims 1 to 3.

Images