Apparatus for controlling vehicle and method thereof

Abstract

The disclosure may relate to a vehicle control apparatus and a method thereof. A vehicle control device may include a sensor, such as a light detection and ranging (LiDAR) sensor, a memory storing a plurality of models, and a processor. The processor may: obtain, via the sensor, a point cloud corresponding to the target object; matching the first reference point with the second reference point on the basis of identifying a target model corresponding to the object type of the target object in the plurality of models; on the basis of matching the first course of the point cloud with the second course of the target model, determining an overlapping proportion with the point cloud in the target model; based on the proportion, determining the occlusion level of the point cloud; and outputting a signal indicative of the occlusion level of the point cloud for controlling the vehicle.

Description

Technical Field

[0001] The present disclosure relates to a vehicle control device and method thereof, and more particularly, to a technology using a sensor such as Light Detection and Ranging (LiDAR). Background Art

[0002] Various studies are being conducted to assist vehicle driving by recognizing external objects using various sensors.

[0003] In particular, when the vehicle is traveling in a driving assist device activation mode (eg, driving assist mode) or an autonomous driving mode, a target object may be recognized by using a sensor (eg, LiDAR).

[0004] The degree to which the target object is blocked needs to be measured by LiDAR. Based on the degree to which the target object is blocked, the process of changing or maintaining the vehicle's driving route can be performed by predicting the movement route of the target object. Summary of the invention

[0005] The present disclosure aims to solve the above-mentioned problems arising in the prior art while maintaining the advantages achieved by the prior art unchanged.

[0006] One aspect of the present disclosure provides a vehicle control apparatus and method thereof, which can identify an occlusion level of a point cloud corresponding to a target object obtained by LiDAR.

[0007] An aspect of the present disclosure provides a vehicle control apparatus and method thereof, which can accurately and quickly identify an occlusion level of a point cloud by identifying the occlusion level of the point cloud based on using a model included in a memory.

[0008] One aspect of the present disclosure provides a vehicle control device and method thereof, which can correct an occlusion level that is incorrectly marked in a point cloud.

[0009] The technical problems to be solved by the present disclosure are not limited to the above-mentioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art to which the present disclosure belongs from the following description.

[0010] According to one or more example embodiments of the present disclosure, a vehicle control device may include: a sensor; a memory storing a plurality of models, each of the plurality of models corresponding to a corresponding object type; and a processor. The processor may be configured to: obtain a point cloud corresponding to a target object via a sensor; match a first reference point contained in the point cloud and corresponding to a specified position of the target object with a second reference point contained in the target model and corresponding to the specified position based on identifying a target model corresponding to the object type of the target object in the plurality of models; and determine a proportion of overlap with the point cloud in the target model based on matching a first heading of the point cloud with a second heading of the target model. Each of the first heading and the second heading may indicate a moving direction of the target object. The processor may also be configured to: determine an occlusion level of the point cloud based on the proportion; and output a signal indicating an occlusion level of the point cloud in order to control the vehicle.

[0011] The processor may be further configured to: train a neural network model based on the point cloud and the occlusion level.

[0012] The processor can be configured to match the first reference point with the second reference point by the following steps: determining a first space containing a first hexahedral form of a point cloud; determining a second space containing a second hexahedral form of a target model; determining a first center point corresponding to an intersection of lines connecting vertices forming the first space, and a second center point corresponding to an intersection of lines connecting vertices forming the second space; and matching the first reference point with the second reference point based on matching the first center point with the second center point.

[0013] The processor may be further configured to match the first heading to the second heading based on scaling the first size of the target model to match a second size of the point cloud.

[0014] The processor may be configured to scale the first size of the target model to match the second size of the point cloud based on changing at least one of a width, a length, and a height of the target model.

[0015] The processor can be configured to determine the scale by the following steps: determining a predetermined horizontal resolution based on the horizontal angular range of the sensor; determining a predetermined vertical resolution based on the vertical angular range of the sensor; segmenting the point cloud into a grid based on the predetermined horizontal resolution and the predetermined vertical resolution; and determining the scale based on the grid.

[0016] The processor may be configured to: determine the scale further by: determining, in each of the point cloud and the target model, voxels segmented by the grid; determining a first shadow region of the point cloud and a second shadow region of the target model; based on identifying a first point corresponding to at least a portion of the target object in at least a portion of the first shadow region, adding a first voxel containing the first point to the first occupied voxel; adding a second voxel containing a second point corresponding to at least a portion of the target object in at least a portion of the second shadow region to the second occupied voxel; and determining the scale based on the first occupied voxel and the second occupied voxel. The first shadow region and the second shadow region may be outside the detection range of the sensor. The first occupied voxel may include a voxel having at least one point identified in the point cloud. The second occupied voxel may include a voxel having at least one point identified in the target model.

[0017] The processor may be further configured to: determine a closest point among the points in the point cloud that is closest to the vehicle; determine a second distance based on applying a predetermined multiplier to the first distance between the vehicle and the closest point; and determine a second occupied voxel based on removing any point from the point cloud that is at least the second distance away from the vehicle.

[0018] The processor may be configured to determine the occlusion level by determining an occlusion level of the point cloud based on a ratio of the first occupied voxels to the second occupied voxels.

[0019] The processor may be further configured to perform labeling on the point cloud based on the occlusion level.

[0020] The processor may be further configured to determine whether to determine an occlusion level based on at least one of a color of the target object and a distance between the vehicle and the target object.

[0021] According to one or more example embodiments of the present disclosure, a vehicle control method may include: obtaining, by a processor, a point cloud corresponding to a target object via a sensor; matching a first reference point included in the point cloud and corresponding to a specified position of the target object with a second reference point included in the target model and corresponding to the specified position based on identifying a target model corresponding to an object type of the target object; and determining a proportion of overlap with the point cloud in the target model based on matching a first heading of the point cloud with a second heading of the target model. Each of the first heading and the second heading may indicate a moving direction of the target object. The method may also include: determining an occlusion level of the point cloud based on the proportion; and outputting a signal indicating the occlusion level of the point cloud in order to control the vehicle.

[0022] The method may also include: training a neural network model based on the point cloud and the occlusion level.

[0023] Matching the first reference point with the second reference point may include: determining a first space containing a first hexahedral shape of a point cloud; determining a second space containing a second hexahedral shape of a target model; determining a first center point corresponding to an intersection of lines connecting vertices forming the first space, and a second center point corresponding to an intersection of lines connecting vertices forming the second space; and matching the first reference point with the second reference point based on matching the first center point with the second center point.

[0024] The method may also include matching the first heading to the second heading based on scaling the first size of the target model to match a second size of the point cloud.

[0025] Scaling may include scaling a first size of the target model to match a second size of the point cloud based on changing at least one of a width, a length, and a height of the target model.

[0026] Determining the scale may include: determining a predetermined horizontal resolution based on a horizontal angular range of the sensor; determining a predetermined vertical resolution based on a vertical angular range of the sensor; segmenting the point cloud into a grid based on the predetermined horizontal resolution and the predetermined vertical resolution; and determining the scale based on the grid.

[0027] Determining the scale may include: determining voxels segmented by the grid in each of the point cloud and the target model; determining a first shadow region of the point cloud and a second shadow region of the target model; adding a first voxel containing the first point to a first occupied voxel based on identifying a first point corresponding to at least a portion of the target object in at least a portion of the first shadow region; adding a second voxel containing a second point corresponding to at least a portion of the target object in at least a portion of the second shadow region to a second occupied voxel; and determining the scale based on the first occupied voxel and the second occupied voxel. The first shadow region and the second shadow region may be outside the detection range of the sensor. The first occupied voxel may include a voxel having at least one point identified in the point cloud. The second occupied voxel may include a voxel having at least one point identified in the target model.

[0028] The method may also include: determining a closest point among the points in the point cloud that is closest to the vehicle; determining a second distance based on applying a predetermined multiplier to the first distance between the vehicle and the closest point; determining a second occupied voxel based on removing any point at least the second distance away from the vehicle from the point cloud; and determining an occlusion level of the point cloud based on a ratio of the first occupied voxel to the second occupied voxel.

[0029] The method may further include determining whether to determine an occlusion level based on at least one of a color of the target object and a distance between the vehicle and the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other objects, features and advantages of the present disclosure will become more apparent through the following detailed description in conjunction with the accompanying drawings:

[0031] Figure 1 An example of a block diagram associated with a vehicle control device according to an embodiment of the present disclosure is shown;

[0032] Figure 2 An example of a model representing an object included in a memory in an embodiment of the present disclosure is shown;

[0033] Figure 3 An example of comparing a point cloud corresponding to a target object with a model corresponding to a type of the target object in an embodiment of the present disclosure is shown;

[0034] Figure 4 An example of determining an occlusion level of a point cloud by using a point cloud and a model in an embodiment of the present disclosure is shown;

[0035] Figure 5 An example of a flowchart associated with a vehicle control method according to an embodiment of the present disclosure is shown;

[0036] Figure 6 shows an example of applying the present disclosure; and

[0037] Figure 7 A computing system related to a vehicle control device or a vehicle control method according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0038] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. When adding reference numerals to the components of each figure, it should be noted that the same components include the same reference numerals even though they are indicated in another figure. In addition, when describing the embodiments of the present disclosure, if detailed descriptions associated with well-known functions or configurations may make the subject matter of the present disclosure unnecessarily obscure, then these detailed descriptions will be omitted.

[0039] When describing the elements of an embodiment of the present disclosure, the terms first, second, A, B, (a), (b), etc. may be used herein. These terms are only used to distinguish one element from another element, but do not limit the corresponding element, regardless of the nature, order or priority of the corresponding element. In addition, unless otherwise defined, all terms including technical or scientific terms used herein should be interpreted as the conventions of the field to which the present disclosure belongs. It should be understood that the terms used herein should be interpreted as including the meanings consistent with their meanings in the context of the present disclosure and the related art, and unless so clearly defined herein, should not be interpreted in an ideal or overly formal sense.

[0040] In the following, reference will be made to Figures 1 to 7 Various embodiments of the present disclosure are described in detail.

[0041] Figure 1 An example of a block diagram associated with a vehicle control device according to an embodiment of the present disclosure is shown.

[0042] Reference Figure 1 The vehicle control device 100 according to the embodiment of the present disclosure may be implemented inside or outside the vehicle, and some components included in the vehicle control device 100 may be implemented inside or outside the vehicle. In this case, the vehicle control device 100 may be integrated with the internal control unit of the vehicle, and may be implemented using a separate device so as to be coupled with the control unit of the vehicle through a separate connection device. For example, the vehicle control device 100 may also include Figure 1 Components not shown.

[0043] The vehicle control device 100 according to the embodiment may include a processor 110, a sensor 120 (e.g., LiDAR 120), and a memory 130. The processor 110, the sensor 120 (e.g., LiDAR 120), or the memory 130 may be electrically and / or operably coupled to each other through electronic components including a communication bus.

[0044] Hereinafter, the fact that a plurality of hardwares are operably coupled may include the fact that a direct and / or indirect connection between the plurality of hardwares is established by wire and / or wirelessly such that a second hardware is controlled by a first hardware among the plurality of hardwares.

[0045] Although various blocks are shown, embodiments are not limited thereto. Figure 1 Some of the multiple hardware in the vehicle control device 100 may be included in a single integrated circuit including a system on a chip (SoC). The type and / or number of hardware included in the vehicle control device 100 is not limited to Figure 1 For example, the vehicle control device 100 may include only Figure 1 Some of the multiple hardware shown in .

[0046] The vehicle control device 100 according to the embodiment may include hardware for processing data based on one or more instructions. The hardware for processing data may include a processor 110. For example, the hardware for processing data may include an arithmetic and logic unit (ALU), a floating point unit (FPU), a field programmable gate array (FPGA), a central processing unit (CPU) and / or an application processor (AP).

[0047] For example, the processor 110 may include a structure of a single-core processor, or may include a structure of a multi-core processor including dual-cores, quad-cores, hexa-cores, or octa-cores.

[0048] The LiDAR 120 included in the vehicle control device 100 according to the embodiment may obtain a data set from identifying objects around the vehicle control device 100. For example, the LiDAR 120 may identify at least one of the position of the surrounding object, the moving direction of the surrounding object, the speed of the surrounding object, or any combination thereof based on a pulse laser signal emitted from the LiDAR 120 being reflected by the surrounding object and returning.

[0049] For example, LiDAR 120 can obtain a data set for expressing external objects in a space defined by a first axis, a second axis, and a third axis based on a pulsed laser signal reflected from surrounding objects. For example, the first axis may include an x-axis. For example, the second axis may include a y-axis. For example, the third axis may include a z-axis. For example, the first axis, the second axis, and the third axis may be perpendicular to each other with the origin as a reference and may intersect each other. The first axis, the second axis, and the third axis are not limited to the above examples. In the following, for convenience of description, the first axis is described as the x-axis; the second axis is described as the y-axis; and the third axis is described as the z-axis.

[0050] For example, the LiDAR 120 may obtain a data set including a plurality of points in a space formed by an x-axis, a y-axis, and a z-axis based on receiving a pulsed laser signal at a specified period.

[0051] The processor 110 included in the vehicle control device 100 according to the embodiment may emit light from the vehicle by using the LiDAR 120. For example, the processor 110 may receive the light emitted from the vehicle. For example, the processor 110 may recognize at least one of the position, speed, moving direction, and any combination thereof of surrounding objects based on the time required to transmit the light emitted from the vehicle and the time required to receive the light emitted from the vehicle.

[0052] For example, the processor 110 can obtain a data set including multiple points based on the time required to send the light emitted from the vehicle and the time required to receive the light emitted from the vehicle. The processor 110 can obtain a data set for expressing multiple points in a three-dimensional virtual coordinate system including an x-axis, a y-axis, and a z-axis.

[0053] The memory 130 included in the vehicle control device 100 according to the embodiment may include a hardware component for storing data and / or instructions to be input and / or output to the processor 110 of the vehicle control device 100 .

[0054] For example, the memory 130 may include a volatile memory including a random access memory (RAM) or a nonvolatile memory including a read only memory (ROM).

[0055] For example, the volatile memory may include at least one of a dynamic RAM (DRAM), a static RAM (SRAM), a cache RAM, or a pseudo SRAM (PSRAM), and any combination thereof.

[0056] For example, the non-volatile memory includes at least one of a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), flash memory, a hard disk, an optical disk, a solid state drive (SSD), an embedded multimedia card (eMMC), and any combination thereof.

[0057] In an embodiment, the memory 130 may include a model representing an object. For example, the model representing an object may include a model representing the object by using a three-dimensional virtual coordinate system. For example, the model representing an object may include a model representing the object by using a plurality of points in a three-dimensional virtual coordinate system.

[0058] In an embodiment, the processor 110 may obtain a point cloud corresponding to the target object through the LiDAR 120. For example, the processor 110 may identify a point cloud corresponding to the target object based on a plurality of points obtained through the LiDAR 120.

[0059] For example, a point cloud may be obtained by performing clustering based on identifying each of a plurality of points obtained by the LiDAR 120 at a specific distance. For example, the point cloud may include a group of points used to create a virtual frame representing a target object.

[0060] In an embodiment, the processor 110 may identify the type of the target object based on the point cloud. For example, the type of the target object may include at least one of a bus, a passenger car, a truck, a sport utility vehicle (SUV), a van, a rubber cone, a marker, and any combination thereof. The type of the target object is not limited to the above examples.

[0061] For example, the processor 110 may identify the type of the target object based on the approximate shape of the point cloud.

[0062] In an embodiment, the processor 110 may identify a model corresponding to the type of the target object among the models representing the objects included in the memory 130 .

[0063] In an embodiment, the processor 110 may identify a first reference point included in the point cloud and corresponding to the specified position of the target object. The processor 110 may identify a second reference point included in the model and corresponding to the specified position of the target object.

[0064] In an embodiment, the processor 110 can match a first reference point contained in the point cloud and corresponding to a specified position of the target object with a second reference point contained in the model and corresponding to the specified position of the target object based on the type of the target object identified by the recognition point cloud and a model corresponding to the type of the target object among the models included in the memory 130.

[0065] In an embodiment, the processor 110 may identify a first heading of the point cloud that indicates a moving direction of the target object.

[0066] In an embodiment, the processor 110 may identify a second heading of the model indicating a moving direction of the target object.

[0067] In an embodiment, the processor 110 may match a first heading of the point cloud indicating the moving direction of the target object with a second heading of the model indicating the moving direction of the target object. For example, the processor 110 may rotate the second heading of the model to match the first heading with the second heading.

[0068] In an embodiment, the processor 110 may identify the ratio of the point cloud to the model based on matching the first heading of the point cloud indicating the moving direction of the target object with the second heading of the model indicating the moving direction of the target object. The ratio may indicate, for example, the ratio between the point cloud and the model relative to the entire model. In other words, the ratio may indicate the ratio (e.g., ratio) of the model that overlaps with the point cloud and the model. For example, a ratio of 30% may indicate that 30% of the model overlaps with the point cloud.

[0069] For example, the processor 110 may identify the ratio of the point cloud in the model based on the matching rate between the first plurality of points included in the identification point cloud and the second plurality of points included in the model.

[0070] In an embodiment, the processor 110 may determine the occlusion level of the point cloud based on the ratio of the point cloud to the model.

[0071] In an embodiment, the point cloud with the occlusion level determined can be used to train a neural network model. For example, the point cloud with the occlusion level determined can be converted into a training data set for training a neural network model. For example, the point cloud with the occlusion level determined can be converted into a training data set for training a neural network model for recognizing external objects for autonomous driving of a vehicle.

[0072] In an embodiment, the processor 110 may express the point cloud in a three-dimensional virtual coordinate system. The processor 110 may segment and identify an area of ​​the point cloud in the three-dimensional virtual coordinate system. For example, the processor 110 may segment and identify an area of ​​the point cloud based on the minimum and maximum values ​​of the x-axis coordinates of the plurality of first points included in the point cloud, the minimum and maximum values ​​of the y-axis coordinates thereof, and the minimum and maximum values ​​of the z-axis coordinates thereof.

[0073] In an embodiment, the processor 110 may identify a first center point of the region (or space) where the point cloud is identified. For example, the first center point of the region (or space) where the point cloud is identified may include an intersection point connecting vertices of a cuboid containing the point cloud in the three-dimensional virtual coordinate system.

[0074] In an embodiment, the processor 110 may identify a second center point of the model. For example, the second center point of the model may include an intersection connecting vertices of a cuboid containing the model expressed in the three-dimensional virtual coordinate system.

[0075] For example, the second center point of the model may be located at (0, 0, 0) in the three-dimensional virtual coordinate system. For example, the model may be normalized to a size of [-1, 1].

[0076] In an embodiment, the processor 110 may identify a first center point corresponding to an intersection point connecting vertices of a first cuboid containing a point cloud. The processor 110 may identify a second center point corresponding to an intersection point connecting vertices of a second cuboid containing a model. The processor 110 may match the first center point with the second center point. The processor 110 may match a first reference point of the point cloud with a second reference point of the model based on matching the first center point with the second center point.

[0077] In an embodiment, the processor 110 may adjust at least one of the first size of the point cloud, the second size of the model, and any combination thereof. For example, the processor 110 may scale the first size of the point cloud to correspond to the second size of the model. For example, the processor 110 may scale the second size of the model to correspond to the first size of the point cloud.

[0078] For example, the processor 110 may change at least one of the width, length, height, and any combination thereof of the model. For example, the processor 110 may change at least one of the width, length, height, and any combination thereof of the point cloud.

[0079] For example, the processor 110 may scale the second size of the model to the first size of the point cloud based on changing at least one of a width, a length, a height, and any combination thereof of the model.

[0080] For example, the processor 110 may scale a first size of the point cloud to a second size of the model based on changing at least one of a width, a length, a height, and any combination thereof of the point cloud.

[0081] In an embodiment, processor 110 may match the first heading to the second heading based on scaling the first size to the second size.

[0082] In an embodiment, processor 110 may match the first heading with the second heading based on scaling the second size to the first size.

[0083] In an embodiment, the processor 110 may identify a pre-specified horizontal resolution and a pre-specified vertical resolution based on the horizontal angle range and the vertical angle range of the LiDAR 120. The processor 110 may segment the point cloud using a grid based on the pre-specified horizontal resolution and the pre-specified vertical resolution. The processor 110 may identify voxels segmented by the grid.

[0084] In an embodiment, the processor 110 may identify the ratio of the point cloud in the model based on the segmented grid. The processor 110 may identify the ratio of the point cloud in the model based on the segmented grid, and may determine the occlusion level of the point cloud (e.g., the occlusion level of the target object) based on the ratio of the point cloud in the model. The processor 110 may control (e.g., move) a vehicle (e.g., a vehicle using the vehicle control device of one or more embodiments of the present disclosure) based on the occlusion level of the target object.

[0085] In an embodiment, the processor 110 may identify a first shadow region (or volume) of the point cloud.The processor 110 may identify a second shadow region (or volume) of the model.

[0086] For example, the first shadow area (or space) or the second shadow area (or space) may include an area (or space) that cannot be observed by the LiDAR 120 .

[0087] In an embodiment, the processor 110 may identify points corresponding to all or a portion of the target object in all or a portion of the first shadow area (or space).

[0088] For example, the processor 110 may add voxels where points corresponding to all or part of the target object exist to the first occupied voxels based on identifying points corresponding to all or part of the target object in all or part of the first shadow region (or space).

[0089] In an embodiment, the processor 110 may identify points representing all or a portion of the target object in all or a portion of the second shadow region (or space).

[0090] For example, processor 110 may add voxels where points representing all or part of the target object exist to the second occupied voxels based on identifying points representing all or part of the target object in all or part of the second shadow region (or space).

[0091] In an embodiment, the processor 110 may identify the shortest point (e.g., the nearest point) closest to the vehicle among the points included in the point cloud. The processor 110 may obtain a second distance obtained by applying a pre-specified ratio to the first distance between the vehicle and the shortest point. The processor 110 may remove points that exist outside the second distance (e.g., points that are at least the second distance away from the vehicle) from the point cloud. The processor 110 may obtain a first occupied voxel based on removing the points that exist outside the second distance from the point cloud.

[0092] In an embodiment, the processor 110 may identify a ratio of a first occupied voxel to a second occupied voxel. For example, the ratio of a first occupied voxel to a second occupied voxel may be referred to as "first occupied voxel / second occupied voxel". The processor 110 may determine an occlusion level of the point cloud based on the ratio of the first occupied voxel to the second occupied voxel.

[0093] For example, the first occupied voxel may include a voxel in which at least one point in the point cloud is identified.

[0094] For example, the second occupied voxels may include voxels where at least one point in the model is identified.

[0095] For example, since the ratio of the first occupied voxel to the second occupied voxel is small, there is almost no occlusion. Since the ratio of the first occupied voxel to the second occupied voxel is large, many occlusions occur.

[0096] In an embodiment, the processor 110 may perform labeling on the point cloud for which the occlusion level has been determined.

[0097] In an embodiment, the processor 110 may determine the occlusion level under a specific condition. For example, the processor 110 may determine whether to determine the occlusion level based on at least one of the color of the target object, the distance between the vehicle and the target object, and any combination thereof.

[0098] For example, the processor 110 may perform a process of determining an occlusion level based on the fact that the color of the target object is a second color different from a first color including black.

[0099] For example, the processor 110 may temporarily stop the process of determining the occlusion level based on the fact that the color of the target object is the first color including black.

[0100] For example, the processor 110 may perform a process of determining an occlusion level based on obtaining a point cloud including a specified number or more of points corresponding to the target object.

[0101] For example, the processor 110 may temporarily stop performing the process of determining the occlusion level based on obtaining a point cloud including points less than a specified number and corresponding to the target object.

[0102] As described above, the processor 110 included in the vehicle control device 100 according to the embodiment can determine the occlusion level of the point cloud based on the point cloud obtained by the LiDAR 120 and the model included in the memory 130. The processor 110 can accurately and quickly determine the occlusion level of the point cloud by determining the occlusion level of the point cloud based on the point cloud and the model. The processor 110 can control (e.g., move) a vehicle (e.g., the vehicle using the vehicle control device of one or more embodiments of the present disclosure) based on the occlusion level of the point cloud representing the target object.

[0103] Figure 2 An example of a model representing an object included in a memory in an embodiment of the present disclosure is shown.

[0104] Reference Figure 2 According to the vehicle control device of the embodiment (eg, Figure 1 A memory (eg, Figure 1 The memory 130 in may include a model 200 representing the object.

[0105] For example, the models 200 may be used to determine the occlusion level of the target object. For example, each model 200 may include type information corresponding to the type of the target object.

[0106] For example, the first model 201 among the models 200 may include a model representing a bus.

[0107] For example, the second model 203 among the models 200 may include a model representing a passenger car.

[0108] For example, the third model 205 among the models 200 may include a model representing at least one of a truck, an SUV, and any combination thereof.

[0109] For example, the fourth model 207 among the models 200 may include a model representing a van.

[0110] For example, the fifth model 209 among the models 200 may include a model representing a rubber cone.

[0111] For example, the sixth model 211 among the models 200 may include a model of an expression marker.

[0112] Figure 2 1 to 6 models 201 to 211 are shown, but the embodiment is not limited thereto.

[0113] In an embodiment, a processor (e.g., Figure 1 The processor 110 may identify a model in the models 200 that corresponds to the type of the target object based on identifying the type of the target object.

[0114] The processor may determine an occlusion level of the target object by using the model based on identifying a model corresponding to the type of the target object.

[0115] For example, the processor may be based on the information that will be obtained by using LiDAR (e.g., Figure 1 The point cloud obtained by the LiDAR 120 in the image processing apparatus and corresponding to the target object is compared with a model corresponding to the type of the target object to determine the occlusion level of the target object.

[0116] For example, the processor may determine an occlusion level of the target object based on comparing a degree of overlap between a point cloud corresponding to the target object and a model corresponding to a type of the target object.

[0117] Hereinafter, a process of determining an occlusion level of a target object will be described later.

[0118] Figure 3 An example of comparing a point cloud corresponding to a target object and a model corresponding to the type of the target object in an embodiment of the present disclosure is shown.

[0119] Reference Figure 3 According to the vehicle control device of the embodiment (eg, Figure 1 A processor (eg, Figure 1 The processor 110 in the embodiment may be configured by LiDAR (e.g., Figure 1 The LiDAR 120 in the image processing apparatus obtains a point cloud corresponding to the target object.

[0120] Reference Figure 3 In example 301 in FIG. 1 , the processor may identify the type of the target object based on a point cloud 303 corresponding to the target object. The processor may match the point cloud 303 with the model 305 based on identifying the model 305 corresponding to the type of the target object.

[0121] For example, the processor may identify a first reference point in point cloud 303 that corresponds to the specified location of the target object. The processor may identify a second reference point in model 305 that corresponds to the specified location of the target object.

[0122] In an embodiment, the processor may match the first reference point with the second reference point based on identifying the first reference point and the second reference point.

[0123] In an embodiment, the processor may identify a first center point of the point cloud 303 contained in the first space. The processor may identify a second center point of the model 305 contained in the second space. For example, the first space may include a hexahedron containing the point cloud 303. For example, the second space may include a hexahedron containing the model 305.

[0124] In an embodiment, the processor may identify a first center point of the first space (e.g., a first center point corresponding to an intersection of lines connecting vertices forming the first space). The processor may identify a second center point of the second space (e.g., a second center point corresponding to an intersection of lines connecting vertices forming the second space).

[0125] In an embodiment, the processor may move the second center point of the model 305 to the origin of the three-dimensional virtual coordinate system. The processor may scale the second size of the model 305 to the first size of the point cloud 303 .

[0126] For example, the processor may change at least one of the width, length, height, and any combination thereof of the model 305. The processor may scale the second size of the model 305 to the first size of the point cloud based on changing at least one of the width, length, height, and any combination thereof of the model 305.

[0127] In an embodiment, the processor may identify the moving direction of the target object. For example, the processor may identify a first heading of the point cloud 303 indicating the moving direction of the target object. For example, the processor may identify a second heading of the model 305 indicating the moving direction of the target object.

[0128] For example, the processor may adjust the first heading of point cloud 303 and the second heading of model 305. For example, the processor may match the second heading of model 305 with the first heading of point cloud 303 by adjusting the second heading of model 305.

[0129] In an embodiment, the processor may identify a distance between a first center point of the point cloud and the three-dimensional virtual coordinate system. The processor may adjust the position of the model based on the distance between the first center point of the point cloud and the three-dimensional virtual coordinate system.

[0130] In an embodiment, the processor may downsample the model based on the distance between the target object and the vehicle. The processor may determine an occlusion level of the point cloud based on downsampling the model.

[0131] In an embodiment, the processor may be configured to detect the presence of a LiDAR (e.g., Figure 1The processor may perform downsampling on the model based on obtaining the point cloud corresponding to the target object by using a sensor different from the LiDAR 120).

[0132] As described above, the processor of the vehicle control device according to the embodiment can adjust the first position of the point cloud and the second position of the model. The processor can determine the occlusion level of the point cloud corresponding to the target object based on the adjustment of the first position and the second position. The processor can accurately determine the occlusion level of the point cloud by determining the occlusion level of the point cloud based on the adjustment of the first position of the point cloud and the second position of the model.

[0133] Figure 4 An example of determining an occlusion level of a point cloud by using a point cloud and a model in an embodiment of the present disclosure is shown.

[0134] Reference Figure 4 According to the vehicle control device of the embodiment (eg, Figure 1 A processor (eg, Figure 1 The processor 110 in the embodiment may obtain a point cloud corresponding to the target object through a sensor 420 (e.g., LiDAR 420). For example, the processor may identify the type of the target object based on obtaining the point cloud corresponding to the target object.

[0135] In an embodiment, the processor may identify the memory (e.g., Figure 1 A model corresponding to the type of the target object among the models included in the memory 130 in the memory.

[0136] In an embodiment, the processor may compare the point cloud and the model based on removing some points included in the model.

[0137] For example, the processor may identify the horizontal angle range and the vertical angle range of the LiDAR 420. For example, the processor may identify a pre-specified horizontal resolution and a pre-specified vertical resolution based on the horizontal angle range and the vertical angle range of the LiDAR 420.

[0138] For example, the processor may segment the point cloud using a grid based on a pre-specified horizontal resolution and a pre-specified vertical resolution.

[0139] For example, the processor may identify a first minimum value 431 corresponding to a minimum value in the horizontal direction. The processor may identify a first maximum value 433 corresponding to a maximum value in the horizontal direction.

[0140] For example, the processor may identify the second minimum value 443 corresponding to the minimum value in the vertical direction. The processor may identify the second maximum value 441 corresponding to the maximum value in the vertical direction.

[0141] In an embodiment, the processor may identify a first line segment extending from the LiDAR 420 to the first minimum 431. The processor may identify a second line segment extending from the LiDAR 420 to the first maximum 433. The processor may identify a third line segment extending from the LiDAR 420 to the second minimum 443. The processor may identify a fourth line segment extending from the LiDAR 420 to the second maximum 441.

[0142] In an embodiment, the processor may identify a first angle 435 between the first line segment and the second line segment. The processor may identify a second angle 445 between the third line segment and the fourth line segment.

[0143] In an embodiment, the processor may segment the point cloud via a grid using a pre-specified horizontal resolution based on a first angle 435 between the first line segment and the second line segment. The processor may segment the point cloud via a grid using a pre-specified vertical resolution based on a second angle 445 between the third line segment and the fourth line segment.

[0144] As described above, the processor may segment the point cloud using a grid according to a pre-specified horizontal resolution based on the first angle 435 and a pre-specified vertical resolution based on the second angle 445 .

[0145] In an embodiment, the processor may identify the ratio of the point cloud to the model based on the segmented mesh. For example, the processor may identify the voxels segmented by the mesh in each of the point cloud and the model.

[0146] For example, the processor may identify the shortest point among the points contained in the point cloud that is closest to the vehicle.

[0147] For example, the processor may identify a first distance between the vehicle and the shortest (eg, closest) point. The processor may obtain a second distance obtained by applying a pre-specified ratio (eg, a predetermined multiplier) to the first distance between the vehicle and the shortest point.

[0148] For example, the processor may remove points that exist outside the second distance (e.g., relative to the vehicle) from the model. For example, because the points that exist outside the second distance are points that exist in an area (or space) that cannot be observed by the LiDAR 420 (e.g., outside the detection range of a sensor such as the LiDAR 420), the processor may remove the points that exist outside the second distance from the model.

[0149] In an embodiment, the processor may identify a first shadow region (or space) of the point cloud and a second shadow region (or space) of the model. For example, the first shadow region (or space) or the second shadow region (or space) may refer to an area (or space) that cannot be observed by LiDAR 420 (e.g., outside the detection range of a sensor such as LiDAR 420).

[0150] For example, the processor may identify a first voxel in which a point corresponding to all or a portion of the target object exists based on identifying points corresponding to all or a portion of the target object in all or a portion of the first shadow region (or space).

[0151] For example, the processor may identify a second voxel in all or a portion of the second shadow region (or space) where there is a point representing all or a portion of the target object.

[0152] In an embodiment, the processor may identify a ratio of the point cloud in the model based on the first voxel and the second voxel.

[0153] In an embodiment, the processor may identify a ratio of the first voxel to the second voxel. The processor may determine an occlusion level of the point cloud based on the ratio of the first voxel to the second voxel.

[0154] As described above, the processor included in the vehicle control device according to the embodiment can obtain the first voxel of the point cloud and the second voxel of the model. The processor can determine the occlusion level of the point cloud based on the first voxel and the second voxel. The processor can accurately identify the occlusion level of the point cloud by determining the occlusion level of the point cloud based on the first voxel and the second voxel. The processor can control (e.g., move) a vehicle (e.g., a vehicle using the vehicle control device of one or more embodiments of the present disclosure) based on the occlusion level of the point cloud (e.g., the occlusion level of the target object).

[0155] Figure 5 An example of a flowchart associated with a vehicle control method according to an embodiment of the present disclosure is shown.

[0156] In the following, it is assumed Figure 1 The vehicle control device 100 executes Figure 5 In addition, Figure 5 In the description, it can be understood that the operations described as being performed by the device are controlled by the processor 110 of the vehicle control device 100.

[0157] Figure 5 At least one of the operations may be Figure 1 The vehicle control device 100 is executed. Figure 5Each of the operations in can be performed sequentially, but need not be performed sequentially. For example, the order of the operations can be changed, and at least two operations can be performed in parallel.

[0158] Reference Figure 5 In operation S501, the vehicle control method according to the embodiment may include: using LiDAR (eg, Figure 1 LiDAR 120 and / or Figure 4 The operation of obtaining a point cloud corresponding to a target object using the LiDAR 420) in the image processing apparatus.

[0159] In operation S503, the vehicle control method according to the embodiment may include: based on identifying the type of the target object identified by the point cloud and a model among the models corresponding to the type of the target object, an operation of matching a first reference point contained in the point cloud and corresponding to a specified position of the target object with a second reference point contained in the model and corresponding to the specified position of the target object.

[0160] The vehicle control method according to the embodiment may include an operation of recognizing a first space including a hexahedral form of a point cloud. The vehicle control method according to the embodiment may include an operation of recognizing a second space including a hexahedral form of a model.

[0161] For example, the vehicle control method may include an operation of identifying a first center point corresponding to an intersection point connecting vertices forming a first space. The vehicle control method may include an operation of identifying a second center point corresponding to an intersection point connecting vertices forming a second space.

[0162] For example, the vehicle control method may include an operation of matching the first center point with the second center point. The vehicle control method may include an operation of matching the first reference point with the second reference point based on matching the first center point with the second center point.

[0163] In operation S505 , the vehicle control method according to the embodiment may include an operation of identifying a ratio of the point cloud in the model based on matching a first heading of the point cloud indicating a moving direction of the target object with a second heading of the model indicating a moving direction of the target object.

[0164] For example, the vehicle control method may include an operation of identifying a first size as a size of the point cloud. For example, the vehicle control method may include an operation of identifying a second size as a size of the model.

[0165] In an embodiment, the vehicle control method may include an operation of matching the first heading with the second heading based on scaling the second size of the model to the first size of the point cloud.

[0166] For example, the vehicle control method may include an operation of scaling a second size of the model to a first size of the point cloud based on changing at least one of a width, a length, a height, and any combination thereof of the model.

[0167] In operation S507 , the vehicle control method according to the embodiment may include an operation of determining an occlusion level of the point cloud based on a ratio of the point cloud in the model.

[0168] For example, a point cloud whose occlusion level is determined can be used to train a neural network model. A neural network model can be trained based on the point cloud and / or the occlusion level.

[0169] The vehicle control method according to the embodiment may include: identifying the operation of the horizontal angle range and the vertical angle range of the LiDAR. For example, the vehicle control method may include: based on the horizontal angle range and the vertical angle range of the LiDAR, identifying the operation of the pre-specified horizontal resolution and the pre-specified vertical resolution.

[0170] For example, the vehicle control method may include an operation of segmenting a point cloud by using a grid based on a pre-specified horizontal resolution and a pre-specified vertical resolution.

[0171] For example, the vehicle control method may include an operation of identifying a ratio of a point cloud in a model based on the segmented mesh.

[0172] The vehicle control method according to the embodiment may include an operation of identifying voxels segmented by a mesh in each of the point cloud and the model.

[0173] For example, the vehicle control method may include an operation of identifying a first shadow area (or space) of the point cloud and a second shadow area (or space) of the model. For example, the first shadow area (or space) or the second shadow area (or space) may include an area (or space) that cannot be observed by the LiDAR.

[0174] For example, the vehicle control method may include: based on identifying points corresponding to all or part of the target object in all or part of the first shadow area (or space), adding voxels in which points corresponding to all or part of the target object exist to the first occupied voxels.

[0175] For example, the vehicle control method may include an operation of adding voxels where points representing all or part of the target object exist in all or part of the second shadow region (or space), to the second occupied voxels.

[0176] For example, the vehicle control method may include an operation of identifying a ratio of the point cloud in the model based on the first occupied voxel and the second occupied voxel.

[0177] The vehicle control method according to the embodiment may include an operation of identifying a shortest point closest to the vehicle among points included in the point cloud.

[0178] For example, the vehicle control method may include an operation of identifying a first distance between the vehicle and the shortest point. The vehicle control method may include an operation of obtaining a second distance obtained by applying a pre-specified ratio (eg, a predetermined multiplier) to the first distance between the vehicle and the shortest point.

[0179] For example, the vehicle control method may include an operation of removing points existing outside the second distance from the model. For example, the vehicle control method may include an operation of obtaining a second occupied voxel based on removing points existing outside the second distance from the model.

[0180] The vehicle control method according to the embodiment may include an operation of determining an occlusion level of the point cloud based on a ratio of the first occupied voxel to the second occupied voxel.

[0181] The vehicle control method may include performing a labeling operation (eg, based on the occlusion level) on the point cloud for which the occlusion level is determined.

[0182] As described above, the vehicle control method may include an operation of determining an occlusion level of the point cloud based on the point cloud corresponding to the target object and a model corresponding to the target object among the models stored in the memory. The vehicle control method may accurately identify the occlusion level of the point cloud by determining the occlusion level of the point cloud.

[0183] Figure 6 An example to which the present disclosure is applied is shown.

[0184] Reference Figure 6 According to the vehicle control device of the embodiment (eg, Figure 1 The vehicle control device 100) can identify the occlusion level of the point cloud corresponding to the target object.

[0185] For example, in Figure 6 In the example, the first text 601 may include an example of determining the occlusion level by applying the present technology. For example, “est_occl:2” in the first text 601 may mean that the occlusion level of the point cloud determined when the present technology is applied is 2.

[0186] Figure 6 The second text 603 in the second text 603 may mean that the ground truth (GT) label is written as 1 before applying the present technology. For example, the "36" included in the second text 603 may include the number of the point cloud corresponding to the target object. For example, the "1" included in the second text 603 may mean that the existing GT label is written as 1.

[0187] Figure 7A computing system related to a vehicle control device or a vehicle control method according to an embodiment of the present disclosure is shown.

[0188] Reference Figure 7 , the computing system 1000 may include at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage 1600 , and a network interface 1700 connected to each other via a bus 1200 .

[0189] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1300 and / or the storage 1600. The memory 1300 and the storage 1600 may include various types of volatile or non-volatile storage media. For example, the memory 1300 may include a ROM (Read Only Memory) 1310 and a RAM (Random Access Memory) 1320.

[0190] Therefore, the process of the method or algorithm described in connection with the embodiments of the present disclosure can be directly implemented by hardware, software modules or combinations thereof executed by the processor 1100. The software module may reside in a storage medium (i.e., memory 1300 and / or storage 1600), such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, solid state drive (SSD), removable disk or CD-ROM. An exemplary storage medium is coupled to the processor 1100, and the processor 1100 can read information from the storage medium and can write information to the storage medium. In another method, the storage medium may be integrated with the processor 1100. The processor 1100 and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In another method, the processor 1100 and the storage medium may reside in a user terminal as separate components.

[0191] Although the present disclosure has been described above with reference to the exemplary embodiments and the accompanying drawings, the present disclosure is not limited thereto but may be variously modified and altered by those skilled in the art without departing from the spirit and scope of the present disclosure as claimed in the following claims.

[0192] Therefore, the exemplary embodiments of the present disclosure are provided to explain the spirit and scope of the present disclosure, rather than to limit it, so that the spirit and scope of the present disclosure are not limited by the embodiments. The scope of the present disclosure should be interpreted based on the appended claims, and all technical ideas within the scope equivalent to the claims should be included in the scope of the present disclosure.

[0193] This technology can identify the occlusion level of the point cloud corresponding to the target object obtained by LiDAR.

[0194] Furthermore, the present technology can accurately and quickly identify the occlusion level of a point cloud by identifying the occlusion level of a point cloud based on using a model included in a memory.

[0195] Furthermore, the present technique can correct for incorrectly labeled occlusion levels in the point cloud.

[0196] Furthermore, various effects directly or indirectly understood through the specification can be provided.

[0197] Although the present disclosure is described above with reference to the exemplary embodiments and the accompanying drawings, the present disclosure is not limited thereto but may be variously modified and altered by those skilled in the art without departing from the spirit and scope of the present disclosure as claimed in the following claims.

Claims

1. A vehicle control device, comprising: sensor; a memory storing a plurality of models, each model of the plurality of models corresponding to a respective object type; and The processor is configured to: obtaining a point cloud corresponding to a target object via the sensor; Based on identifying a target model corresponding to the object type of the target object among the multiple models, matching a first reference point included in the point cloud and corresponding to a specified position of the target object with a second reference point included in the target model and corresponding to the specified position; determining a proportion of the point cloud in the target model based on matching a first heading of the point cloud with a second heading of the target model, wherein each of the first heading and the second heading indicates a moving direction of the target object; determining an occlusion level of the point cloud based on the ratio; and In order to control the vehicle, a signal indicative of an occlusion level of the point cloud is output.

2. The vehicle control device according to claim 1, wherein: The processor is further configured to: A neural network model is trained based on the point cloud and the occlusion level.

3. The vehicle control device according to claim 1, wherein: The processor is configured to: The first reference point is matched with the second reference point by the following steps: Determining a first space of a first hexahedral shape containing the point cloud; determining a second space containing a second hexahedral shape of the target model; determining a first center point corresponding to an intersection point of lines connecting vertices forming the first space, and a second center point corresponding to an intersection point of lines connecting vertices forming the second space; as well as Based on matching the first center point with the second center point, the first reference point is matched with the second reference point.

4. The vehicle control device according to claim 1, wherein: The processor is further configured to: The first heading is matched to the second heading based on scaling a first size of the object model to match a second size of the point cloud.

5. The vehicle control device according to claim 4, wherein: The processor is configured to: A first size of the target model is scaled to match a second size of the point cloud based on changing at least one of a width, a length, and a height of the target model.

6. The vehicle control device according to claim 1, wherein: The processor is configured to: The ratio is determined by the following steps: Determining a predetermined horizontal resolution based on a horizontal angle range of the sensor; Determining a predetermined vertical resolution based on a vertical angular range of the sensor; Segmenting the point cloud into a grid based on the predetermined horizontal resolution and the predetermined vertical resolution; as well as Based on the grid, the scale is determined.

7. The vehicle control device according to claim 6, wherein: The processor is configured to: The ratio is also determined by the following steps: In each of the point cloud and the object model, determining voxels segmented by the grid; Determining a first shadow area of ​​the point cloud and a second shadow area of ​​the target model; based on identifying a first point corresponding to at least a portion of the target object in at least a portion of the first shadow region, adding a first voxel containing the first point to the first occupied voxels; adding a second voxel including a second point corresponding to at least a portion of the target object in at least a portion of the second shadow region to the second occupied voxel; as well as determining the ratio based on the first occupied voxel and the second occupied voxel, wherein the first shadow area and the second shadow area are outside the detection range of the sensor, wherein the first occupied voxel comprises a voxel having at least one point identified in the point cloud, and Wherein, the second occupied voxels include voxels having at least one point identified in the object model.

8. The vehicle control device according to claim 7, wherein: The processor is further configured to: Determining a closest point among the points in the point cloud that is closest to the vehicle; determining a second distance based on applying a predetermined multiplier to the first distance between the vehicle and the closest point; as well as The second occupied voxel is determined based on removing any point from the point cloud that is at least the second distance away from the vehicle.

9. The vehicle control device according to claim 7, wherein: The processor is configured to: The occlusion level is determined by the following steps: An occlusion level of the point cloud is determined based on a ratio of the first occupied voxels to the second occupied voxels.

10. The vehicle control device according to claim 1, wherein: The processor is further configured to: Based on the occlusion level, labeling is performed on the point cloud.

11. The vehicle control device according to claim 1, wherein: The processor is further configured to: Whether to determine the occlusion level is determined based on at least one of a color of the target object and a distance between a vehicle and the target object.

12. A vehicle control method comprising the following steps: A processor obtains a point cloud corresponding to the target object via a sensor; Based on identifying a target model corresponding to an object type of the target object, matching a first reference point included in the point cloud and corresponding to a specified position of the target object with a second reference point included in the target model and corresponding to the specified position; determining a proportion of the point cloud in the target model based on matching a first heading of the point cloud with a second heading of the target model, wherein each of the first heading and the second heading indicates a moving direction of the target object; Based on the ratio, determining an occlusion level of the point cloud; and In order to control the vehicle, a signal indicative of an occlusion level of the point cloud is output.

13. The method according to claim 12, further comprising the steps of: A neural network model is trained based on the point cloud and the occlusion level.

14. The method according to claim 12, wherein: The step of matching the first reference point with the second reference point comprises: Determining a first space of a first hexahedral shape containing the point cloud; determining a second space containing a second hexahedral shape of the target model; determining a first center point corresponding to an intersection point of lines connecting vertices forming the first space, and a second center point corresponding to an intersection point of lines connecting vertices forming the second space; and Based on matching the first center point with the second center point, the first reference point is matched with the second reference point.

15. The method according to claim 12, further comprising the steps of: The first heading is matched to the second heading based on scaling a first size of the object model to match a second size of the point cloud.

16. The method according to claim 15, wherein: The scaling steps include: A first size of the target model is scaled to match a second size of the point cloud based on changing at least one of a width, a length, and a height of the target model.

17. The method according to claim 12, wherein: The steps of determining the ratio include: Determining a predetermined horizontal resolution based on a horizontal angle range of the sensor; Determining a predetermined vertical resolution based on a vertical angular range of the sensor; Segmenting the point cloud into grids based on the predetermined horizontal resolution and the predetermined vertical resolution; and Based on the grid, the scale is determined.

18. The method according to claim 17, wherein: The steps of determining the ratio include: In each of the point cloud and the object model, determining voxels segmented by the grid; Determining a first shadow area of ​​the point cloud and a second shadow area of ​​the target model; based on identifying a first point corresponding to at least a portion of the target object in at least a portion of the first shadow region, adding a first voxel containing the first point to the first occupied voxels; adding a second voxel including a second point corresponding to at least a portion of the target object in at least a portion of the second shadow region to the second occupied voxel; and determining the ratio based on the first occupied voxel and the second occupied voxel, The first shadow area and the second shadow area are outside the detection range of the sensor. wherein the first occupied voxel comprises a voxel having at least one point identified in the point cloud, and Wherein, the second occupied voxels include voxels having at least one point identified in the object model.

19. The method according to claim 18, further comprising the steps of: Determining a closest point among the points in the point cloud that is closest to the vehicle; determining a second distance based on applying a predetermined multiplier to the first distance between the vehicle and the closest point; determining the second occupied voxel based on removing from the point cloud any point that is at least the second distance away from the vehicle; as well as An occlusion level of the point cloud is determined based on a ratio of the first occupied voxels to the second occupied voxels.

20. The method according to claim 12, further comprising the steps of: Whether to determine the occlusion level is determined based on at least one of a color of the target object and a distance between a vehicle and the target object.

Images