Vehicle control device

Abstract

To provide a vehicle control device capable of reducing processing time and a processing load when determining whether an object is located on a road or off the road.SOLUTION: A vehicle control device that provides a driving support function by detecting an object around an own vehicle comprises: an object recognition section that recognizes the object as point cloud data based on distance information detected by a distance sensor detecting a distance to the object; a storage section that stores map data and grid data obtained by dividing the map data into a plurality of grids of a predetermined size, and including grids classified as on-road and grids classified as off-road; a specifying section that specifies the grids to which each grid data of a portion of the points in the point cloud data recognized by the object recognition section belongs; and a determination section that determines whether the object corresponding to the point cloud data exists on the road or off the road based on the classification of the grids to which each of a portion of the points specified by the specifying section belong.SELECTED DRAWING: Figure 3

Description

Technical Field

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

Background Art

[0002] In recent years, research has been underway on autonomous driving and driving support functions that allow a vehicle to travel without depending on the driver's operation. As a control technology for such autonomous driving and driving support functions, for example, there is disclosed a technology related to a vehicle control device that detects an object using a lidar mounted on a vehicle and identifies whether the detected object is present around the vehicle based on the detected point cloud data (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the conventional technology, when an object is present near the boundary between outside and inside the road, in order to determine whether the object is present inside or outside the road using the entire point cloud data of the object obtained from a lidar or the like, there is a problem that a large amount of processing time and processing load are required.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a vehicle control device capable of reducing the processing time and processing load when determining whether an object is present inside or outside the road.

Means for Solving the Problems

[0006] In order to solve the above-described problems and achieve the object, a vehicle control device according to the present invention is a vehicle control device that detects an object around a host vehicle and provides a driving support function, and includes an object recognition unit that recognizes the object as point cloud data based on distance information detected by a distance measurement sensor that detects a distance to the object, map data, and grid data obtained by dividing the map data into a plurality of grids of a predetermined size and including at least grids classified outside a road and grids classified inside a road, a specifying unit that specifies to which grid of the grid data each of a part of the points of the point cloud data recognized by the object recognition unit belongs, and a determining unit that determines whether the object corresponding to the point cloud data exists inside a road or outside a road based on the classification of the grids to which the part of the points specified by the specifying unit belong.

Effects of the Invention

[0007] According to the present invention, it is possible to reduce the processing time and processing load when determining whether an object exists inside a road or outside a road.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments of the vehicle control device according to the present invention will be described in detail with reference to FIGS. 1 to 8. Also, the present invention is not limited by the following embodiments, and the constituent elements in the following embodiments include those that can be easily conceived by those skilled in the art, those that are substantially the same, and those within the so-called equivalent range. Furthermore, various omissions, substitutions, changes, and combinations of the constituent elements can be made without departing from the gist of the following embodiments.

[0010] (Regarding the System Configuration of the Vehicle) FIG. 1 is a diagram showing an example of the system configuration of a vehicle according to an embodiment. FIG. 2 is a diagram for explaining a case where a target exists inside the road, outside the road, and partially outside the road. With reference to FIGS. 1 and 2, the system configuration and operation outline of the vehicle 1 according to the present embodiment will be described.

[0011] The vehicle 1 shown in FIG. 1 has an automatic driving function and is a vehicle capable of traveling by automatic driving without depending on the driving operation of the user (driver). Note that the automatic driving includes semi-automatic driving in which a part of the operations in the traveling of the vehicle 1 is automated (requiring partial driving operations by the user). These automatic driving and semi-automatic driving functions correspond to the "driving support function" of the present invention.

[0012] The vehicle 1 is equipped with a plurality of ECUs (Electronic Control Units) to control each part. Each ECU includes a microcomputer (Micro Controller Unit), and the microcomputer incorporates, 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).

[0013] As shown in FIG. 1, the vehicle 1 includes, as the above-mentioned ECUs, a drive ECU 11, a steering ECU 12, a brake ECU 13, a meter ECU 14, a body ECU 15, a drive device 21, a steering device 22, a braking device 23, and an emergency stop SW 24. The drive ECU 11, the steering ECU 12, the brake ECU 13, the meter ECU 14, the body ECU 15, and an automatic driving ECU 31 described later are connected so as to enable communication (CAN communication) according to the CAN (Controller Area Network) communication protocol.

[0014] The drive ECU 11 is an ECU that controls the drive device 21 of the vehicle 1. The drive device 21 is a drive device having at least one of an engine or a motor as a drive source. The drive device 21 includes a transmission that shifts and outputs the driving force from the drive source as necessary.

[0015] The steering ECU 12 is an ECU that controls the steering device 22 of the vehicle 1. The steering device 22 is, for example, an electric power steering device that applies the torque of an electric motor to a steering mechanism. The steering mechanism includes, for example, a rack and pinion type steering gear, and is configured such that when the rack shaft moves in the vehicle width direction by the torque of the electric motor, the left and right steering wheels steer left and right as the rack shaft moves.

[0016] The brake ECU 13 is a control unit that controls the braking device 23 of the vehicle 1. The braking device 23 is a hydraulic or electric braking device. In the case of a hydraulic type, the braking device 23 includes a brake actuator, and by the function of this brake actuator, hydraulic pressure is distributed to the wheel cylinders of the brakes provided on each wheel, and braking force is applied to the wheels including the drive wheels by this hydraulic pressure.

[0017] The meter ECU 14 is an ECU that controls each part of the meter panel (not shown) of the vehicle 1. The meter panel includes instruments that display the vehicle speed and engine speed, and a display such as a liquid crystal display for displaying various information. Further, an emergency stop SW 24 that is operated to instruct an emergency stop of the automatic driving is connected to the meter ECU 14.

[0018] The body ECU 15 is an ECU that controls each part that needs to operate even when the ignition switch of the vehicle 1 is off, for example, the left and right winkers and door lock motors.

[0019] As shown in FIG. 1, the vehicle 1 further includes an automatic driving ECU 31, a LiDAR (Light Detection And Ranging) ECU 32, a monocular camera ECU 33, an omnidirectional LiDAR 34, a positioning signal receiver 35, a LiDAR 36, a monocular camera 37, and an HMI (Human Machine Interface) device 38.

[0020] The automatic driving ECU 31 is an ECU that performs control of the automatic driving control. The automatic driving ECU 31 is an example of a vehicle control device. As shown in FIG. 1, the automatic driving ECU 31 is communicably connected to the drive ECU 11, the steering ECU 12, the brake ECU 13, the meter ECU 14, and the body ECU 15 via CAN communication. Further, the LiDAR ECU 32, the monocular camera ECU 33, the omnidirectional LiDAR 34, the positioning signal receiver 35, and the HMI device 38 are communicably connected to the automatic driving ECU 31.

[0021] For automatic driving control, the automatic driving ECU 31 recognizes targets such as other vehicles present around the vehicle 1 using the information detected by the omnidirectional radar 34 and the monocular camera 37. Here, Fig. 2 shows the position of the target TG, which is another vehicle or the like, relative to the road. Fig. 2(a) shows that the target TG exists within the road, that is, the whole of the target TG exists inside the road line RL indicating the road edge such as the outer road line. Fig. 2(b) shows that the target TG exists outside the road, that is, the whole of the target TG exists outside the road line RL. Fig. 2(c) shows that the target TG exists partially outside the road, that is, the target TG exists on the road line RL. Hereinafter, the road line shall be a line indicating the road edge such as the outer road line. In the automatic driving by the automatic driving ECU 31, it is necessary to appropriately recognize whether the target, which is another vehicle or the like, exists within the road, outside the road, or on the road line (partially outside the road), and when it exists within the road or on the road line, it is necessary to cause avoidance or stop or the like. In this case, if the recognition process is performed using, for example, all of the point cloud data obtained from the omnidirectional radar 34 or the like, the processing time and processing load will be increased. For the vehicle 1 equipped with the automatic driving ECU 31 according to the present embodiment, the details of the operation for reducing the processing time and processing load in the process (road inside / outside determination process) for determining whether the target exists inside or outside the road will be described below.

[0022] The lidar ECU 32 is an ECU that controls the operations of a plurality of lidars 36. For example, six lidars 36 are connected to the lidar ECU 32. The lidar ECU 32 is communicably connected to the autonomous driving ECU 31 via a communication cable such as the Ethernet (registered trademark) standard or the USB (Universal Serial Bus) standard. The lidar 36 is a device that detects the distance from the vehicle 1 to an object by irradiating a laser beam in the search range and detecting the reflected light from the object. The lidar 36 outputs the detected distance information to the lidar ECU 32, and the lidar ECU 32 transmits the processed data regarding the distance information to the autonomous driving ECU 31. The lidars 36 are respectively arranged, for example, at the left end, the center, and the right end of the front bumper of the vehicle 1, and at the left end, the center, and the right end of the rear bumper.

[0023] The monocular camera ECU 33 is an ECU that controls the operation of the monocular camera 37. The monocular camera ECU 33 has the monocular camera 37 connected thereto. The monocular camera ECU 33 is communicably connected to the autonomous driving ECU 31 via a communication cable such as the Ethernet standard or the USB standard. The monocular camera 37 is an imaging device capable of continuously capturing still images of the search range in front of the vehicle 1 at a predetermined frame rate. The monocular camera 37 outputs the image data of the continuously captured still images to the monocular camera ECU 33, and the monocular camera ECU 33 transmits the processed data regarding the image data to the autonomous driving ECU 31.

[0024] The omnidirectional lidar 34 is a device that irradiates laser light in all 360° directions and detects the distance from the vehicle 1 to an object present within the search range by detecting the reflected light from the object. The omnidirectional lidar 34 is communicably connected to the autonomous driving ECU 31 via a communication cable such as the Ethernet standard or the USB standard, and outputs the detected distance information to the autonomous driving ECU 31. The distance information can take the form of, for example, point cloud data indicating the distance from the vehicle 1 to an object at each position (voxel) in a three-dimensional space. Note that the omnidirectional lidar 34 corresponds to the "distance measurement sensor" of the present invention.

[0025] The positioning signal receiver 35 is a receiving device that receives positioning signals from positioning satellites based on GNSS (Global Navigation Satellite System). The positioning signal receiver 35 is communicably connected to the autonomous driving ECU 31 via a communication cable such as a USB standard, and outputs the received positioning signal to the autonomous driving ECU 31. The autonomous driving ECU 31 detects the position where the vehicle 1 is located based on the positioning signal received from the positioning signal receiver 35. As an example of GNSS, for example, GPS (Global Positioning System) etc. can be mentioned.

[0026] The HMI device 38 is a display device disposed in the passenger compartment of the vehicle 1 that displays map information and object recognition information. The HMI device 38 is communicably connected to the autonomous driving ECU 31 via a communication cable such as an Ethernet standard or a USB standard.

[0027] Note that, instead of the omnidirectional lidar 34 and the lidar 36, a millimeter-wave radar or the like may be used.

[0028] (Configuration and operation of the functional blocks of the vehicle's autonomous driving ECU) FIG. 3 is a diagram showing an example of the configuration of the functional blocks of the autonomous driving ECU of the vehicle according to the embodiment. FIG. 4 is a diagram for explaining the operation of converting the sensor coordinates of the vehicle and the target in the embodiment into map coordinates. FIG. 5 is a diagram for explaining high-precision map data and grid data. FIG. 6 is a diagram for explaining the processing when the point of the target data exists in the grid on the road line in the vehicle according to the embodiment. With reference to FIGS. 3 to 6, the configuration and operation of the functional blocks of the autonomous driving ECU 31 of the vehicle 1 according to the present embodiment will be described.

[0029] As shown in FIG. 3, the autonomous driving ECU 31 includes an object recognition unit 41, a self-position estimation unit 42, a surrounding information integration unit 43, a route planning unit 44, a vehicle control unit 45, and a storage unit 46.

[0030] The object recognition unit 41 is a functional unit that recognizes the target objects (obstacles such as other vehicles, pedestrians, buildings, curbs, etc.) existing around the vehicle 1 from the distance information to the target objects detected by the omnidirectional lidar 34 and the image captured by the monocular camera 37.

[0031] Specifically, the object recognition unit 41 performs a ground deletion process of deleting the point group indicating the ground from the point cloud based on the distance information detected by the omnidirectional lidar 34. Further, the object recognition unit 41 performs a clustering process of grouping the point clouds for each point cloud with a close distance between the point clouds. Furthermore, the object recognition unit 41 performs a boxing process of fitting the grouped point clouds into a rectangular parallelepiped shape in order to recognize the target objects existing around the vehicle 1 which is the host vehicle. Then, the object recognition unit 41 performs a tracking process of tracking while calculating the relative position and relative speed of the rectangular parallelepiped-shaped target object subjected to the boxing process with respect to the vehicle 1. Thereby, the object recognition unit 41 can recognize the target objects. Note that a boxing process of fitting the grouped point clouds into a rectangular parallelepiped shape is performed, and the information of the point cloud to be the tracking target by the tracking process is referred to as three-dimensional point cloud data.

[0032] The self-position estimation unit 42 is a functional unit that estimates the position (self-position) of the vehicle 1 by matching the distance information to the target objects detected by the omnidirectional lidar 34 and the high-precision three-dimensional map data (high-precision map data) stored in the storage unit 46. In this case, the coordinates of the position of the vehicle 1 estimated by the self-position estimation unit 42 are the coordinates on the high-precision map data (hereinafter, may be referred to as map coordinates). The high-precision map data includes, for example, points indicating the positions of roads, information on road lines connecting the points, information such as the width and gradient of the roads, road markings, shoulder lines, intersections, level crossings, stop lines, crosswalks, signs, and other ground features, and information on structures derived from road structures. Note that the self-position estimation unit 42 may integrate the self-position estimated by the matching of the distance information and the high-precision map data and the self-position based on the positioning signal received by the positioning signal receiver 35 to improve the estimation accuracy of the self-position. Further, the high-precision map data corresponds to the "map data" of the present invention.

[0033] The peripheral information integration unit 43 is a functional unit that inputs the recognition result of the target by the object recognition unit 41, the estimation result of the self-position by the self-position estimation unit 42, and the data processed from the distance information output from each lidar 36 by the lidar ECU 32, and integrates the peripheral information. For example, the peripheral information integration unit 43 creates peripheral information integration map data in which the vehicle 1, vehicles other than the vehicle 1, and targets such as pedestrians are arranged on the high-precision map data. Then, the peripheral information integration unit 43 outputs and displays the high-precision map data or the peripheral information integration map data, etc., and the recognition information of the target on the HMI device 47.

[0034] As shown in FIG. 3, the peripheral information integration unit 43 includes a generation unit 431, a conversion unit 432, a specification unit 433, a road inside / outside determination unit 434, a size determination unit 435, and a determination unit 436. Note that the road inside / outside determination unit 434 corresponds to the "determination unit" of the present invention.

[0035] The generation unit 431 is a functional unit that generates two-dimensional point cloud data obtained by removing the data of the coordinates in the height direction (Z coordinates) from the three-dimensional point cloud data of the target recognized by the object recognition unit 41. For example, in order to generate two-dimensional point cloud data, the generation unit 431 may extract the point cloud data corresponding to the bottom surface or the top surface of the three-dimensional point cloud data in the shape of a rectangular parallelepiped as the two-dimensional point cloud data, or the point cloud data composed of the plane coordinates obtained by removing the coordinates in the height direction of all the points of the three-dimensional point cloud data may be used as the two-dimensional point cloud data. That is, the two-dimensional point cloud data generated by the generation unit 431 becomes rectangular point cloud data. Note that the two-dimensional point cloud data generated by the generation unit 431 may be simply referred to as point cloud data hereinafter. As a result, the data for processing by the automatic driving ECU 31 can be reduced, and the processing time and processing load can be reduced.

[0036] Note that since the high-precision map data includes structures derived from road structures as described above, the generation unit 431 may exclude the three-dimensional point cloud data corresponding to the structures derived from the road structures indicated by the high-precision map data from the three-dimensional point cloud data of the object recognized by the object recognition unit 41. This can reduce the processing time and processing load.

[0037] The conversion unit 432 is a functional unit that converts the coordinates of the point cloud data of the object generated by the generation unit 431 into map coordinates on the high-precision map data based on the position of the vehicle 1 estimated by the self-position estimation unit 42. Since the point cloud data of the object generated by the generation unit 431 is based on the distance information detected by the omnidirectional lidar 34 as described above, as shown in FIG. 4(a), the point cloud data of the object TG recognized by the object recognition unit 41 is data indicating coordinates (hereinafter sometimes referred to as sensor coordinates) with the position of the omnidirectional lidar 34 as a reference (origin). Therefore, as shown in FIG. 4(b), the conversion unit 432 converts the sensor coordinates of the object into map coordinates on the high-precision map data in order to process the point cloud data of the object on the high-precision map data.

[0038] The specifying unit 433 is a functional unit that specifies to which grid in the grid data each point of the point cloud data converted into map coordinates by the conversion unit 432 belongs. That is, the specifying unit 433 specifies to which row and column of the grid in the grid data each point of the point cloud data belongs. Here, the grid data refers to data obtained by dividing high-precision map data into grids of a predetermined size (for example, grids of 1×1 [m] square) in a matrix form. As shown in FIG. 5(a), the map information MI, which is high-precision map data, includes points indicating the positions of roads and information on road lines connecting these points. And a part of the grid data obtained by dividing the map information MI shown in FIG. 5(a) into grids of a predetermined size is shown as the grid data GD in FIG. 5(b). Values indicating outside the road (for example, 0), values indicating on the road line (for example, 1), and values indicating inside the road (for example, 2) are assigned to each grid of the grid data. That is, each grid included in the grid data is classified into one of outside the road, inside the road, and on the road line. Therefore, as shown in FIG. 5(b), the grid data GD includes an area G1 that is an area of grids to which values indicating outside the road are assigned, an area G2 that is an area of grids to which values indicating inside the road are assigned, and an area G3 that is an area of grids to which values indicating on the road line are assigned. That is, the area G1 outside the road and the area G2 inside the road are separated by the area G3 of the road line. Such grid data is assumed to be created in advance and stored in the storage unit 46 in advance. In this way, by creating the grid data in advance and storing it in the storage unit 46, no processing time and processing load for separately creating the grid data occur during the processing by the automatic driving ECU 31. Note that the grid data may be data different from the high-precision map data or may be data integrated with the configuration degree map data.

[0039] The road inside / outside determination unit 434 is a functional unit that determines whether each point of the point cloud data specified by the specifying unit 433 is inside or outside the road based on the value of the grid to which each point belongs. For example, when the value of the grid to which the point targeted by the point cloud data belongs indicates inside the road, the road inside / outside determination unit 434 determines that the point is inside the road, and when the value of the grid indicates outside the road, the road inside / outside determination unit 434 determines that the point is outside the road. Also, the processing when the value of the grid to which the point targeted by the point cloud data belongs indicates on the road line will be described with reference to FIG. 6.

[0040] As shown in FIG. 6, in principle, a value indicating the road line is assigned to the grid through which the road line included in the high-precision map data passes. In FIG. 6, the road point RP1 passing through the road points RP1 and RP2 is shown. Here, when the four vertices of the rectangular point group data indicating the target are respectively the vertices T1, T2, T3, and T4, for example, it is assumed that a value indicating the road line is assigned to the grid FG corresponding to the vertex T1. In this case, the road inside / outside determination unit 434 first determines whether one side is inside the road or outside the road and whether the other side is outside the road or inside the road with the road line RL shown in FIG. 6 as the boundary based on the values of the grids adjacent to the grid FG to which the vertex T1 to be determined belongs in the vertical and horizontal directions. In the example shown in FIG. 6, since the grids adjacent to the upper side and the left side of the grid FG to which the vertex T1 belongs are grids indicating outside the road, the road inside / outside determination unit 434 determines that the upper side (left side) of the road line RL is outside the road and the lower side (right side) of the road line RL is inside the road. Next, the road inside / outside determination unit 434 determines on which side of the road line RL the vertex T1 to be determined is located. In the example shown in FIG. 6, the road inside / outside determination unit 434 determines that the vertex T1 is located on the upper side (left side) of the road line RL. Then, since the vertex T1 is located on the upper side (left side) of the road line RL and the upper side (left side) of the road line RL is outside the road, the road inside / outside determination unit 434 determines that the vertex T1 is outside the road. That is, when the grid to which the point included in the point group data specified by the specifying unit 433 belongs is classified on the road line, if the point is located inside the road with the road line in the map data as the boundary, the road inside / outside determination unit 434 determines that the point is inside the road, and if the point is located outside the road with the road line as the boundary, the road inside / outside determination unit 434 determines that the point is outside the road.

[0041] In the example shown in FIG. 6, values indicating on the road line are also assigned to the grids corresponding to vertices T2 and T3. However, by the same processing as described above by the road inside / outside determination unit 434, it is determined that vertex T2 is inside the road and vertex T3 is outside the road. Thus, even when the grid to which the points included in the point cloud data belong is a grid on the road line, it is possible to accurately determine whether the point is inside the road or outside the road.

[0042] The size determination unit 435 is a functional unit that determines whether the size of the object indicated by the point cloud data converted into map coordinates by the conversion unit 432 is equal to or greater than the vehicle size.

[0043] The determination unit 436 is a functional unit that determines whether the object indicated by the point cloud data converted into map coordinates by the conversion unit 432 exists inside the road or outside the road based on the determination results by the road inside / outside determination unit 434 and the size determination unit 435.

[0044] The route planning unit 44 is a functional unit that plans a travel route from the peripheral information integrated map data generated by the peripheral information integration unit 43 to the destination of the vehicle 1. The travel route includes the driving route of the vehicle 1 and the target vehicle speed at each point on the driving route. The route planning unit 44 causes the HMI device 38 to display the route data of the planned travel route.

[0045] The vehicle control unit 45 is an ECU that outputs commands to control the operations 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 by automatic driving according to the travel route planned by the route planning unit 44. For example, after the destination of the vehicle 1 is input via the HMI device 38 and the automatic driving start button displayed on the HMI device 38 is pressed, an instruction to start automatic driving is input from the HMI device 38 to the automatic driving ECU 31. When an instruction to start automatic driving is input to the automatic driving ECU 31, the route planning unit 44 plans a travel route to the destination of the vehicle 1. The travel route is replanned at a predetermined cycle by the route planning unit 44 during the automatic driving of the vehicle 1.

[0046] The storage unit 46 is a functional unit that stores the above-described high-precision map data and grid data. The storage unit 46 is realized by a non-volatile storage device such as a flash memory, for example.

[0047] Then, the vehicle 1 travels at the target vehicle speed for each point on the driving route included in the driving route according to the most newly planned driving route. The automatic driving ends, for example, when the vehicle 1 arrives at the destination or when the emergency stop SW24 is pressed and an instruction to stop the automatic driving is input from the meter ECU 14 to the automatic driving ECU 31.

[0048] Note that each functional unit of the automatic driving ECU 31 shown in FIG. 3 conceptually shows the functions and is not limited to such a configuration. For example, a plurality of functional units illustrated as independent functional units in the automatic driving ECU 31 shown in FIG. 3 may be configured as one functional unit. On the other hand, the functions of one functional unit in the automatic driving ECU 31 shown in FIG. 3 may be divided into a plurality of parts and configured as a plurality of functional units. Further, each functional unit of the automatic driving ECU 31 does not necessarily need to be configured as a clear software module as the block shown in FIG. 3, and it is sufficient that the functions of each functional unit are realized as a whole when a program is executed in the automatic driving ECU 31.

[0049] (Flow of Vehicle's Road Inside / Outside Determination Process) FIG. 7 is a flowchart showing an example of the flow of the vehicle's road inside / outside determination process according to the embodiment. FIG. 8 is a diagram for explaining the determination process when the target crosses the road line in the vehicle's road inside / outside determination process according to the embodiment. The flow of the vehicle 1's road inside / outside determination process according to the present embodiment will be described with reference to FIGS. 7 and 8.

[0050] <Step S11> First, the generation unit 431 of the automatic driving ECU 31 generates two-dimensional point cloud data by removing the data of the coordinates in the height direction (Z coordinates) from the three-dimensional point cloud data of the target recognized by the object recognition unit 41. Then, it proceeds to step S12.

[0051] <Step S12> The conversion unit 432 of the automatic driving ECU 31 converts the coordinates of the four vertices at the four corners of the point cloud data of the target generated by the generation unit 431 into map coordinates on the high-precision map data based on the position of the vehicle 1 estimated by the self-position estimation unit 42. Then, it proceeds to step S13.

[0052] <Step S13> The identification unit 433 of the automatic driving ECU 31 identifies which grid in the grid data each of the four vertices of the point cloud data converted into map coordinates by the conversion unit 432 belongs to. Note that the points to be identified regarding which grid the point cloud data belongs to are not limited to the four vertices as described above. As long as some points in the point cloud data are sufficient for determining whether it is inside or outside the road in the next step S14, they can be used. Then, it proceeds to step S14.

[0053] <Step S14> The road inside / outside determination unit 434 of the automatic driving ECU 31 determines whether each of the four vertices of the point cloud data identified by the identification unit 433 is inside or outside the road based on the values of the grids to which the four vertices belong. When the value of the grid to which the vertex belongs indicates on the road line, it is determined whether the vertex is inside or outside the road by the process described in FIG. 6 above. Then, it proceeds to step S15.

[0054] <Step S15> As a result of the determination by the road inside / outside determination unit 434, if all of the four vertices of the point cloud data are inside the road (step S15: Yes), the process proceeds to step S24, and if at least any one of the four vertices is outside the road (step S15: No), the process proceeds to step S16.

[0055] <Step S16> As a result of the determination by the road inside / outside determination unit 434, if all of the four vertices of the point cloud data are outside the road (step S16: Yes), the process proceeds to step S25, and if at least any one of the four vertices is inside the road (step S16: No), the process proceeds to step S17.

[0056] <Step S17> The size determination unit 435 of the automatic driving ECU 31 determines whether the size of the object indicated by the point cloud data converted into map coordinates by the conversion unit 432 is equal to or greater than the vehicle size. At this time, since the four vertices of the point cloud data include both vertices inside the road and vertices outside the road, it is determined that the object indicated by the point cloud data straddles the road line. If the size of the object indicated by the point cloud data is equal to or greater than the vehicle size (step S17: Yes), the process proceeds to step S18, and if it is less than the vehicle size (step S17: No), the process proceeds to step S26.

[0057] <Step S18> The conversion unit 432 converts the coordinates of all the points of the point cloud data of the object into map coordinates on the high-precision map data based on the position of the vehicle 1 estimated by the self-position estimation unit 42. Then, the process proceeds to step S19.

[0058] <Step S19> The specifying unit 433 specifies, for each of all the points of the point cloud data converted into map coordinates by the conversion unit 432, which grid in the grid data it belongs to. Then, the process proceeds to step S21.

[0059] <Step S20> Based on the values of each grid to which all the points of the point cloud data specified by the specifying unit 433 belong, the road inside / outside determination unit 434 determines whether each of all the points is inside the road or outside the road. When the value of the grid to which each point belongs indicates on the road line, it is determined whether each of the points is inside the road or outside the road by the process described in FIG. 6 above. Then, the process proceeds to step S21.

[0060] <Step S21> Then, the road inside / outside determination unit 434 determines whether or not more than 50% of all the points of the point cloud data are inside the road. Here, FIG. 8 shows an example of the determination operation of whether the point cloud data of the target is inside the road or outside the road. Since all the points of the point cloud data of the target TG1 indicated by "A" shown in FIG. 8 are outside the road, it is determined that more than 50% of the points are not inside the road. Since all the points of the point cloud data of the target TG2 indicated by "B" are inside the road, it is determined that more than 50% of the points are inside the road. And since more than half of the points constituting the point cloud data of the target TG3 indicated by "C" are outside the road, it is determined that more than 50% of the points are not inside the road. When more than 50% of the points are inside the road (step S21: Yes), the process proceeds to step S22. When less than 50% of the points are inside the road (step S21: No), the process proceeds to step S23. Note that the determination is not limited to whether or not more than 50% of the points are inside the road, and it may be a determination of whether or not more than a predetermined ratio of the points are inside the road. For example, if it is possible to consider that the target is inside the road when more than 30% of the points are inside the road, the predetermined ratio may be set to 30% for determination.

[0061] <Step S22> When more than 50% of all the points of the point cloud data are inside the road, the determination unit 436 of the automatic driving ECU 31 determines that the target indicated by the point cloud data exists inside the road. Then, the road inside / outside determination process ends.

[0062] <Step S23> If less than 50% of all the points in the point cloud data are within the road, the determination unit 436 determines that the object indicated by the point cloud data exists outside the road. Then, the road inside / outside determination process ends.

[0063] <Step S24> If all four vertices of the point cloud data are within the road, the determination unit 436 determines that the object indicated by the point cloud data exists within the road. Then, the road inside / outside determination process ends.

[0064] <Step S25> If all four vertices of the point cloud data are outside the road, the determination unit 436 determines that the object indicated by the point cloud data exists outside the road. Then, the road inside / outside determination process ends.

[0065] <Step S26> If the size of the object indicated by the point cloud data is smaller than the vehicle size, since the object indicated by the point cloud data is determined to be a road structure, vegetation, etc., the determination unit 436 determines that the object indicated by the point cloud data exists outside the road. Then, the road inside / outside determination process ends.

[0066] As described above, the automatic driving ECU 31 of the vehicle 1 determines whether the point cloud data indicating the object exists within the road or outside the road, and excludes the object existing outside the road from the identification targets in automatic driving, etc., thereby suppressing the occurrence of misrecognition and the occurrence of misstops and the like due to the misrecognition.

[0067] Note that the above road inside / outside determination process is executed for each object recognized by the object recognition unit 41.

[0068] (Effect of this Embodiment) As described above, in the automatic driving ECU 31 of the vehicle 1 according to the present embodiment, the object recognition unit 41 recognizes a target as point cloud data based on the distance information detected by the omnidirectional lidar 34 that detects the distance to the target, and the storage unit 46 stores the high-precision map data and grid data obtained by dividing the high-precision map data into a plurality of grids of a predetermined size and including at least grids classified outside the road and grids classified inside the road. The specifying unit 433 specifies to which grid of the grid data each of a part of the points of the point cloud data recognized by the object recognition unit 41 belongs, and the determining unit 436 determines whether the target corresponding to the point cloud data exists inside the road or outside the road based on the classification of the grids to which the part of the points specified by the specifying unit 433 belong. As a result, it is not necessary to analyze the coordinates of each point of the point cloud data, and since the classification of the grids in the grid data is used, the processing time and processing load for determining whether the target exists inside the road or outside the road can be reduced.

[0069] Also, in the automatic driving ECU 31 of the vehicle 1 according to the present embodiment, the point cloud data is rectangular point cloud data, and the specifying unit 433 specifies to which grid of the grid data each of the four vertices of the point cloud data belongs. As a result, it is not necessary to use all of the point cloud data and only the four vertices are used, so the processing time and processing load for determining whether the target exists inside the road or outside the road can be reduced.

[0070] In addition, in the automatic driving ECU 31 of the vehicle 1 according to the present embodiment, when a part of the points specified by the specifying unit 433 straddles the road line indicating the road edge in the classification of the grid to which each belongs, the road inside / outside determination unit 434 determines whether or not, among all the points included in the point cloud data, a predetermined ratio or more of the points are inside the road. When it is determined by the road inside / outside determination unit 434 that a predetermined ratio or more of all the points included in the point cloud data are inside the road, the determination unit 436 determines that the target corresponding to the point cloud data exists inside the road. When it is determined that the ratio is less than the predetermined ratio of the points are inside the road, it is determined that the target corresponding to the point cloud data exists outside the road. As a result, for example, it is possible to suppress misrecognition as a target existing inside the road when, for example, vegetation planted outside the road extends and protrudes into the road, and it is possible to suppress the vehicle 1 from stopping erroneously.

[0071] In addition, in the automatic driving ECU 31 of the vehicle 1 according to the present embodiment, the grid data is composed of grids classified outside the road, grids classified inside the road, and grids classified on the road line indicating the road edge. When the grid to which the points included in the point cloud data specified by the specifying unit 433 belong is classified on the road line, if the point is located inside the road with the road line in the high-precision map data as a boundary, the road inside / outside determination unit 434 determines that the point is inside the road. If the point is located outside the road with the road line as a boundary, the road inside / outside determination unit 434 determines that the point is outside the road. Based on the determination result by the road inside / outside determination unit 434, the determination unit 436 determines whether the target corresponding to the point cloud data exists inside the road or outside the road. As a result, even when the points of the point cloud data belong to a grid classified on the road, it is possible to accurately determine whether the point is outside the road or inside the road, so that the accuracy of determining whether the target is outside the road or inside the road can be improved.

Explanation of Signs

[0072] 1 Vehicle 11 Driving ECU 12 Steering ECU 13 Brake ECU 14-meter ECU 15-body ECU 31-autopilot ECU 32-rider ECU 33-monocular camera ECU 34-omnidirectional lidar 35-positioning signal receiver 36-lidar 37-monocular camera 38-HMI device 41-object recognition unit 42-ego position estimation unit 43-surrounding information integration unit 44-route planning unit 45-vehicle control unit 46-memory unit 431-generation unit 432-conversion unit 433-specification unit 434-road inside / outside determination unit 435-size determination unit 436-decision unit

Claims

1. A vehicle control device that detects an object around a vehicle and provides a driving support function, an object recognition unit that recognizes the object as point cloud data based on distance information detected by a distance measuring sensor that detects the distance to the object; a storage unit that stores map data and grid data obtained by dividing the map data into a plurality of grids of a predetermined size and including at least grids classified outside the road and grids classified inside the road; a specifying unit that specifies to which grid of the grid data each of a part of the points of the point cloud data recognized by the object recognition unit belongs; a determination unit that determines whether the object corresponding to the point cloud data exists inside the road or outside the road based on the classification of the grids to which the part of the points specified by the specifying unit belong; A vehicle control device comprising:

2. The point cloud data is rectangular point cloud data, The vehicle control device according to claim 1, wherein the specifying unit specifies to which grid of the grid data each of the four vertices of the point cloud data belongs as the part of the points.

3. When the object straddles a road line indicating a road edge in a state where the grids to which the part of the points specified by the specifying unit belong are classified, a determination unit that determines whether or not a predetermined ratio or more of all the points included in the point cloud data are inside the road is further provided, The determination unit determines that the object corresponding to the point cloud data exists inside the road when it is determined that a predetermined ratio or more of all the points included in the point cloud data are inside the road by the determination unit, and determines that the object corresponding to the point cloud data exists outside the road when it is determined that the ratio is less than the predetermined ratio. The vehicle control device according to claim 1 or 2.

4. The grid data is composed of grids classified outside the road, grids classified inside the road, and grids classified on a road line indicating a road edge, When the grid to which the points included in the point cloud data specified by the specifying unit belong is classified on the road line, a determination unit that determines that the point is inside the road when the point is located inside the road with the road line in the map data as a boundary, and determines that the point is outside the road when the point is located outside the road with the road line as a boundary is further provided, The vehicle control device according to claim 1 or 2, wherein the determination unit determines whether the target corresponding to the point cloud data exists inside or outside the road based on the determination result by the determination unit.

Images