Device for controlling a vehicle and method therefor

Abstract

A vehicle control device may include a light detection and ranging device (LiDAR) and a processor. The processor may identify a first speed of an external vehicle based on identifying a first virtual box corresponding to the external vehicle. The processor may determine a confidence level of the first speed and whether the first speed is stable. If the first speed is determined to be unstable, the processor may assign one of a plurality of speeds to a second virtual box.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a vehicle control device and method, and more particularly to a technology for identifying an external object using one or more sensors, such as light detection and ranging devices (LiDAR). BACKGROUND

[0002] An external object can be identified using various sensors to assist a vehicle in driving.

[0003] In particular, while driving in a driver assistance system activation mode or in an autonomous driving mode, the vehicle can identify the external object using a sensor, e.g. a LiDAR device.

[0004] The speed of an external vehicle may not be accurately measured by the LiDAR. While the vehicle is driving in driver assistance system activation mode or in autonomous driving mode, driving safety and the driving experience can be improved if the speed of the external vehicle can be measured more accurately. SUMMARY

[0005] The following summary is a simplified summary of certain features. The summary is not a comprehensive overview and is not intended to identify important or critical elements.

[0006] One aspect of the present disclosure provides a vehicle control device that identifies a speed to be assigned to a virtual box using at least one of a speed of an external vehicle and a displacement of the virtual box corresponding to the external vehicle, or any combination thereof, and a method therefor.

[0007] One aspect of the present disclosure provides a vehicle control device that determines a virtual box having an unstable speed and assigns a stable speed to the virtual box having the unstable speed, and a corresponding method.

[0008] One aspect of the present disclosure provides a vehicle control device that reduces a speed error caused by a shape error of an object, and a corresponding method.

[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 here will be clearly apparent from the following description to those skilled in the art to which the present disclosure refers.

[0010] A device of a first vehicle may comprise: at least one sensor; and a processor, the processor configured to: determine, based on the at least one sensor, a first virtual box corresponding to an external vehicle and associated with a first time; identify, based on the first virtual box, a first speed of the external vehicle at the first time; determine, based on information of the first virtual box, information of a second virtual box identified at a second time after the first time, or driving information of the first vehicle, or any combination thereof, whether a confidence level of the first speed exceeds a reference confidence level;determine whether the first speed is stable based on a first displacement obtained by a difference between a first position of the first virtual box and a second position of the second virtual box, a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time point, and a predetermined threshold displacement after the confidence level of the first speed exceeds the reference confidence level;selecting a second displacement having the smallest difference from the predicted displacement from among displacements based on a difference between first representative points included in the first virtual box and second representative points, each corresponding to the first representative points, included in the second virtual box, after determining the first velocity as unstable; and assigning to the second virtual box a second velocity based on the selected second displacement, a third velocity based on the second displacement and a third displacement obtained by applying an algorithm to first points included in the first virtual box and second points included in the second virtual box, or a fourth velocity based on the second displacement and the predicted displacement;

[0011] The processor may be configured to: identify the confidence level of the first speed based on at least one of the following: a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the first vehicle identified at the first time and a second yaw rate of the first vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are in a boundary region of the at least one sensor, or whether the first virtual box or the second virtual box is obstructed, or any combination thereof.The boundary region of the at least one sensor may comprise a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor.

[0012] The processor may be configured to: increase the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box being less than or equal to a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box being less than or equal to a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box being less than or equal to a reference width, the first virtual box and the second virtual box not being in the boundary region of the at least one sensor, a component of a first axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box being greater than a component of a second axis direction,the first virtual box and the second virtual box are not obscured or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time is less than or equal to a reference yaw rate.

[0013] The processor may be configured to: determine the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box exceeding a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box exceeding a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box exceeding a reference width, the difference between a speed of the first vehicle at the first time and the speed of the first vehicle at the second time exceeding a reference speed, or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time,which exceeds a reference yaw rate.

[0014] The processor may be configured to: determine that the first speed is unstable based on a difference between the first displacement and the predicted displacement exceeding the threshold displacement; and determine that the first speed is stable based on the difference between the first displacement and the predicted displacement being less than or equal to the threshold displacement.

[0015] The processor may be configured to: assign the first velocity to the second virtual box based on the confidence level of the first velocity being less than or equal to a first reference confidence level; determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a first threshold displacement when the confidence level of the first velocity exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level;Determining that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is smaller than the first threshold displacement when the confidence level of the first velocity exceeds the second reference confidence level; and determining that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement that exceeds the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary region of the at least one sensor. The boundary region of the at least one sensor may comprise a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor.

[0016] The processor may be configured to: identify, as at least one of the first representative points and the second representative points, at least one of the vertices of the first virtual box and the second virtual box, a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, and points contained in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis.

[0017] The processor may be configured to: determine at least one of the first representative points or at least one of the second representative points based on a direction in which the first virtual box is located relative to the first vehicle, based on the first virtual box or the second virtual box identified in a boundary region of the at least one sensor. The boundary region of the at least one sensor may include a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor.

[0018] The processor may be configured to: assign the second speed to the second virtual box based on identifying the first virtual box and the second virtual box as exceeding a certain distance from the first vehicle, so as not to apply the algorithm to the first points and the second points.

[0019] The processor may be configured to: obtain the third displacement based on applying the algorithm to the first points and the second points, based on the first virtual box and the second virtual box being identified at a certain distance or less from the first vehicle to apply the algorithm to the first points and the second points; and assign the third speed based on an average of the third displacement and the second displacement to the second virtual box.

[0020] The processor may be configured to: determine a fourth displacement by applying a weight to each of the second displacement and the predicted displacement based on a difference between the second displacement and the predicted displacement that is greater than or equal to a threshold; and assign the fourth velocity to the second virtual box based on determining the fourth velocity using the fourth displacement.

[0021] These and other features and benefits are described in more detail below. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings: Fig. 1 shows an example of a block diagram of a vehicle control device; Fig. 2 shows an example of moving a virtual box; Fig. Figure 3 shows an example of identifying representative points of a virtual box; Fig. Figure 4 shows an example in which a virtual box is identified in a boundary area of ​​a LiDAR; Fig. 5 shows an example of determining a displacement based on first representative points of a virtual box identified at a first time and second representative points of a virtual box identified at a second time; Fig. 6 shows an example of a result to which one or more aspects of the present disclosure are applied and a comparative example of a result to which the one or more aspects of the present disclosure are not applied; Fig.7 shows an example of a flowchart relating to a vehicle control method; Fig. 8 shows an example of a flowchart relating to a vehicle control method; and Fig. 9 illustrates an example of a computer system associated with a vehicle control device or method. DETAILED DESCRIPTION

[0023] Below, various examples of the present disclosure will be described in detail with reference to the accompanying drawings. When identifying components in each drawing, it should be noted that the same components have the same reference numerals even if they are shown in a different drawing. Furthermore, in describing the features of the present disclosure, detailed descriptions associated with well-known functions or configurations are omitted when they might unnecessarily obscure the subject matter of the present disclosure.

[0024] In describing elements of the present disclosure, the terms first, second, A, B, (a), (b), and the like may be used herein. These terms are used only to distinguish one element from another and do not limit the corresponding elements, regardless of the type, order, or priority of the corresponding elements. Furthermore, unless otherwise defined, all terms used herein, including technical and scientific terms, are to be interpreted as is customary in the art to which the present disclosure belongs.It is to be understood that the terms used herein should be construed to have a meaning consistent with their meaning in the context of the present disclosure and the relevant prior art, and should not be construed in an idealized or overly formal sense unless expressly so defined herein.

[0025] Below, various examples of the present disclosure are described with reference to the Fig. 1 to 9 described in detail.

[0026] Fig. 1 shows an example of a block diagram of a vehicle control device.

[0027] Referring to Fig.1, a vehicle control device 100 may be implemented inside or outside a vehicle, and some of the components included in the vehicle control device 100 may be implemented inside or outside the vehicle. In some embodiments, the vehicle control device 100 may be integrated into internal control units of a vehicle and implemented with a separate device for connecting to control units of the vehicle using a separate connector and / or connection interface. The vehicle control device 100 may, for example, include further Fig. 1 components not shown included.

[0028] Referring to Fig.1, the vehicle control device 100 may include a processor 110 and one or more sensors (e.g., a LiDAR 120 and / or other object detection sensors capable of transmitting a signal and receiving the signal reflected from an object). The processor 110 or the LiDAR 120 may be electrically and / or operatively connected to each other by an electronic component including a communication bus.

[0029] Hereinafter, operatively coupled hardware parts may comprise hardware components that are coupled via a direct and / or indirect connection between the hardware parts and may be implemented wired and / or wirelessly such that the second hardware is controlled by the first hardware among the hardware parts.

[0030] Although various blocks are shown, the aspects are not limited to them. Some of the Fig.1 may be included in a single integrated circuit, including a "System on a Chip" (SoC). The type or number of hardware included in the vehicle control device 100 is not limited to the Fig. 1. For example, the vehicle control device 100 may only implement some of the Fig. 1 parts of the hardware shown.

[0031] The vehicle control device 100 may include hardware for processing data based on one or more instructions. The data processing hardware may include the processor 110. The data processing hardware may include, for example, 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). The processor 110 may have the structure of a single-core processor or a multi-core processor, including a dual-core, quad-core, hexa-core, or octa-core processor, etc.

[0032] The LiDAR 120 of the vehicle control device 100 can obtain data sets from the identification of objects located in the vicinity of the vehicle control device 100. For example, the LiDAR 120 can identify at least one of the object's location, a direction of movement of the object, a speed of the object, or any combination thereof, for example, based on a pulsed laser signal emitted by the LiDAR 120 that is reflected and returned by the object.

[0033] For example, based on a pulsed laser signal reflected from the object(s), the LiDAR 120 may obtain data sets including a plurality of points of the object(s) in the space defined by a first axis, a second axis, and a third axis. For example, the LiDAR 120 may determine data sets including a plurality of points in the space defined by the first axis, the second axis, and the third axis based on receiving the pulsed laser signal with a specific period. The first axis may include, for example, an x-axis (e.g., the heading direction of the vehicle, the longitudinal axis of the vehicle, or the direction from the rear center of the vehicle to the front center of the vehicle). The second axis may include, for example, a y-axis (e.g., the horizontal axis of the vehicle, the transverse axis of the vehicle). The third axis may include, for example, a z-axis (e.g., the vertical axis of the vehicle).However, the first axis, the second axis and / or the third axis are not limited to the examples above.

[0034] The processor 110 included in the vehicle control device 100 can emit light from a vehicle using the LiDAR 120. For example, the processor 110 can receive light emitted by the vehicle through the LiDAR 120. For example, the processor 110 can identify at least one of an object's location, speed, or direction of movement, or any combination thereof, based, for example, on a time taken to transmit light emitted by the vehicle and a time taken to receive light emitted by the vehicle. A LiDAR device can measure the time it takes for the emitted light to travel to an object and back. The LiDAR device can record the time between the time the laser pulse leaves the LiDAR device and the time it returns to calculate the distance between the LiDAR device and the object.

[0035] For example, the processor 110 may obtain data sets including a plurality of points based on the time required to transmit light emitted by the vehicle and the time required to receive light emitted by the vehicle. The processor 110 may obtain data sets expressing a plurality of points in a three-dimensional virtual coordinate system including the first axis, the second axis, and the third axis.

[0036] The processor 110 may identify a first virtual box corresponding to an external vehicle via the LiDAR 120. For example, the processor 110 may identify the first virtual box based on a first data set acquired at a first time. For example, the processor 110 may identify the first virtual box based on a plurality of points obtained at the first time and included in the first data set.

[0037] In one example, the processor 110 may identify, via the LiDAR 120, a first speed of the external vehicle at the first time based on the first virtual box corresponding to the external vehicle.

[0038] In one example, processor 110 may identify at least one of the first virtual box information or vehicle driving information including the vehicle controller 100, or any combination thereof. Processor 110 may determine whether a confidence level of the first speed exceeds a reference confidence level based on at least one of the first virtual box information or vehicle driving information, or any combination thereof.

[0039] For example, the first virtual box information may include at least one of the following information: a size of the first virtual box, a first direction of the first virtual box, a shape of the first virtual box, whether the first virtual box is located within a boundary region of the LiDAR 120, whether the first virtual box is occluded, or any combination thereof. For example, the boundary region of the LiDAR 120 may include a region within a certain distance from the boundary line of the field of view (FOV) of the LiDAR 120.

[0040] The driving information of the vehicle may include, for example, at least a yaw rate of the vehicle, an indication of whether the vehicle is traveling on a ramp section, whether the vehicle is rotating, or a rate of change of speed of the vehicle, or any combination thereof.

[0041] For example, the processor may determine the confidence level of the first speed based on Table 1 below. [Table 1] Increasing the confidence level Condition +n 1) If the difference between the size of the first virtual box and the size of the second virtual box is less than or equal to the reference rate 2) If the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box is less than or equal to the reference angle 3) If the difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to the reference width 4) If the first virtual box and the second virtual box are not located in the boundary region of the LiDAR 120 5) If the component of the first axis direction is greater than the component of the second axis direction contained in the first heading direction of the first virtual box and the second heading direction of the second virtual box 6) If the first virtual box and the second virtual box are not covered, or 7) If the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time is less than or equal to the reference yaw rate +n / 2 1) If the first virtual box or the second virtual box is hidden 2) If the difference between the first yaw rate and the second yaw rate exceeds the reference yaw rate, but the difference between the speed of the vehicle detected at the first time and the speed of the vehicle detected at the second time is less than or equal to the reference speed Reset 1) If the confidence level exceeds 'm', the confidence level is changed to 'k' if the conditions mentioned under 2) are met 2) If the confidence level exceeds 'k' and is less than 'm' a. If the difference between the first width of the first virtual box and the second width of the second virtual box exceeds the reference width b. If the first virtual box or the second virtual box is identified in the boundary region of the LiDAR 120 c. If the component of the second axis direction contained in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than the component of the first axis direction d. If the difference between the first position of the first virtual box and the second position of the second virtual box exceeds the reference distance e. If the difference between the speed of the vehicle at the first time and the speed of the vehicle at the second time exceeds the threshold

[0042] In one example, the processor 110 may increase the confidence level of the first speed by 'n' based on the fact that a difference between the size of the first virtual box and the size of the second virtual box is less than or equal to a reference rate, a difference between the first heading direction of the first virtual box and a second heading direction of the second virtual box is less than or equal to a reference angle, a difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to a reference width, the first virtual box and the second virtual box are not located in a boundary region of the LiDAR 120, a component of the first axis direction is greater than a component of the second axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box,the first virtual box and the second virtual box are not obscured, or a difference between a first yaw rate of the vehicle identified at the first time and a second yaw rate of the vehicle identified at the second time is less than or equal to a reference yaw rate.

[0043] In one example, the processor 110 may increase the confidence level of the first speed by “n / 2” based on the fact that the first virtual box or the second virtual box is occluded or a difference between the first yaw rate and the second yaw rate exceeds a reference yaw rate, but a difference between the speed of the vehicle identified at the first time and the speed of the vehicle identified at the second time is less than or equal to a reference speed.

[0044] In one example, the processor 110 may initialize the confidence level of the first speed or change the confidence level of the first speed to a specific value based on the fact that a difference between a first width of the first virtual box and a second width of the second virtual box exceeds a reference width, the first virtual box or the second virtual box is identified in a boundary region of the LiDAR 120, a component of the second axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than a component of the first axis direction,a difference between a first location of the first virtual box and a second location of the second virtual box exceeds a reference distance or a difference between the speed of the vehicle at the first time and the speed of the vehicle at the second time exceeds a threshold value.

[0045] In one example, processor 110 may determine that the confidence level of the first velocity exceeds the reference confidence level. Processor 110 may identify a first position of the first virtual box at the first time. Processor 110 may identify a second position of the second virtual box at the second time. If the confidence level of the first velocity exceeds the reference confidence level, processor 110 may determine a difference between the first location of the first virtual box and the second location of the second virtual box. Processor 110 may determine a first displacement by the difference between the first position and the second position.

[0046] In one example, processor 110 may identify a predicted displacement of the first virtual box by the first speed of the external vehicle at the first time. For example, processor 110 may obtain the predicted displacement of the first virtual box by using the first speed and a difference between a time at which the first virtual box is obtained and a time at which the second virtual box is obtained.

[0047] In one example, processor 110 may determine whether the first velocity is stable based on the first displacement, the predicted displacement, and a predetermined threshold displacement. For example, the threshold displacement may be set to a value that varies depending on the confidence level of the first velocity.

[0048] In one example, the processor 110 may assign the first speed to the second virtual box if the confidence level of the first speed is less than or equal to the reference confidence level.

[0049] In one example, if the confidence level of the first velocity exceeds the reference confidence level, the processor 110 may determine the first displacement based on the difference between the first position of the first virtual box and the second position of the second virtual box.

[0050] In one example, processor 110 may predict the predicted displacement of the first virtual box based on the first velocity of the external vehicle at the first time. For example, processor 110 may predict the predicted displacement of the first virtual box based on applying the first velocity of the external vehicle at the first time to the first virtual box.

[0051] In one example, when the confidence level of the first speed exceeds the reference confidence level, the processor 110 may determine whether the first speed is stable based on the first displacement obtained by a difference between the first position of the first virtual box and the second position of the second virtual box, the predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time, and a predetermined threshold displacement.

[0052] For example, if the confidence level of the first velocity exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level, the processor 110 may determine that the first velocity is unstable based on a difference between the first displacement and the predicted displacement exceeding a first threshold displacement.

[0053] For example, if the confidence level of the first velocity exceeds the second reference confidence level, the processor 110 may determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is less than the first threshold displacement.

[0054] For example, if the first virtual box is identified in a boundary region of a LiDAR, the processor 110 may determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement that exceeds the first threshold displacement.

[0055] For example, the first reference confidence level may contain approximately 5. The second reference confidence level may contain approximately 20. The first threshold shift may contain approximately 0.1 m, for example. The second threshold shift may contain approximately 0.05 m, for example. The third threshold shift may contain approximately 0.2 m, for example.

[0056] In one example, processor 110 may identify first representative points contained within the first virtual box. Processor 110 may identify second representative points contained within the second virtual box.

[0057] For example, if it is determined that the first velocity is unstable, the processor 110 may identify displacements based on a difference between first representative points included in the first virtual box and second representative points, each corresponding to the first representative points and included in the second virtual box.

[0058] For example, if it is determined that the first velocity is unstable, the processor 110 may select the second displacement having the smallest difference from the predicted displacement from the displacements based on the difference between the first representative points included in the first virtual box and the second representative points, each corresponding to the first representative points and included in the second virtual box.

[0059] For example, the processor 110 may identify vertices of the first virtual box as at least one of the first representative points. For example, the processor 110 may identify at least one of a first axis minimum value, a first axis maximum value, a second axis minimum value, or a second axis maximum value, or any combination thereof, among the points contained in the first virtual box on a plane formed by a first axis and a second axis among the first axis, the second axis, and the third axis.The processor 110 may identify, as at least one of the first representative points of the first virtual box, at least one of the minimum value of the first axis, the maximum value of the first axis, the minimum value of the second axis, or the maximum value of the second axis, or any combination thereof, among the points included in the first virtual box on the plane formed by the first axis and the second axis among the first axis, the second axis, and the third axis.

[0060] For example, the processor 110 may identify vertices of the second virtual box as at least one of the second representative points. For example, the processor 110 may identify at least one of a first axis minimum value, a first axis maximum value, a second axis minimum value, or a second axis maximum value, or any combination thereof, among the points contained in the second virtual box on the plane formed by the first axis and the second axis, among the first axis, the second axis, and the third axis.The processor 110 may identify as at least one of the second representative points of the second virtual box at least one of the points included in the second virtual box on the plane formed by the first axis and the second axis between the first axis, the second axis, and the third axis: the minimum value of the first axis, the maximum value of the first axis, the minimum value of the second axis, or the maximum value of the second axis, or any combination thereof.

[0061] In one example, when the first virtual box or the second virtual box is identified in a boundary region of the LiDAR 120, the processor 110 may identify at least one of some of the first representative points or some of the second representative points, or any combination thereof, based on a direction in which the first virtual box is located relative to the vehicle. Descriptions of the first representative points and the second representative points are provided later in the Fig. described.

[0062] In one example, processor 110 may determine a second velocity based on the selected second displacement. For example, processor 110 may determine the second velocity based on dividing a second displacement by a time difference between the first time and the second time.

[0063] For example, if the first virtual box and the second virtual box are identified beyond a certain distance (e.g., approximately 80 m) from the vehicle, the processor 110 may not apply an algorithm to the first points and the second points. For example, if the first virtual box and the second virtual box are identified beyond the specified distance from the vehicle so as not to apply the algorithm to the first points and the second points, the processor 110 may assign the second speed to the second virtual box.

[0064] In one example, the processor 110 may apply the algorithm to the first points in the first virtual box and the second points in the second virtual box. The algorithm may, for example, include an iterative closest point (ICP) algorithm. The ICP algorithm may, for example, include an algorithm for estimating a correspondence relationship between the first points and the second points. The processor 110 may obtain a third displacement by applying the algorithm to the first points in the first virtual box and the second points in the second virtual box. The processor 110 may obtain a third velocity based on dividing the third displacement by the time difference between the first time and the second time.

[0065] In one example, if the first virtual box and the second virtual box are identified at a certain distance (e.g., approximately 80 m) or less from the vehicle, the processor 110 may apply the algorithm to the first points and the second points. If the first virtual box and the second virtual box are identified at a certain distance or less from the vehicle to apply the algorithm to the first points and the second points, the processor 110 may determine the third displacement based on applying the algorithm to the first points and the second points. The processor 110 may determine the average of the third offset and the second offset. The processor 110 may determine a third speed based on the average of the third displacement and the second displacement.For example, the processor may determine the third speed by dividing the average of the third displacement and the second displacement by the time difference between the first time and the second time. Processor 110 may assign the third speed, based on the average of the third displacement and the second displacement, to the second virtual box.

[0066] In one example, processor 110 may determine a fourth velocity based on the second displacement and the predicted displacement. For example, processor 110 may determine a fourth displacement based on applying weights to the second displacement and the predicted displacement. For example, processor 110 may determine a fourth displacement that is half the sum of the second displacement to which the first weight is applied and the predicted displacement to which the second weight is applied by applying a first weight to the second displacement and applying a second weight to the predicted displacement. Processor 110 may determine the fourth velocity using the fourth path.For example, processor 110 may determine the fourth velocity based on dividing the fourth displacement by the time difference between the first time and the second time. Processor 110 may assign the fourth velocity to the second virtual box based on determining the fourth velocity using the fourth displacement.

[0067] As described above, when the first speed of the external vehicle identified at the first time is unstable, the processor 110 of the vehicle control device 100 may perform various operations to stabilize the speed of the second virtual box identified at the second time. The processor 110 may assign a second speed to the second virtual box based on the selected second displacement, a third speed based on the second displacement and the third displacement obtained by applying the algorithm to the first points in the first virtual box and the second points in the second virtual box, or a fourth speed based on the second displacement and the predicted displacement, thereby stabilizing a speed of the second virtual box identified at the second time.

[0068] Fig. 2 shows an example of moving a virtual box.

[0069] With reference to Fig. 2, a processor (e.g., processor 110 in Fig. 1) a vehicle control device (e.g. the vehicle control device 100 of Fig. 1) obtain at least a first displacement 225 or a predicted displacement 215 or any combination thereof based on the virtual boxes 211 and 221 of a first time point.

[0070] Referring to a first example 201 in Fig. 2, the processor may identify a first virtual box 211 at the first time. The processor may identify a first location of the first virtual box 211 at the first time. The processor may determine a first speed 217 of an external vehicle identified at the first time.

[0071] For example, the processor may predict a second position of a second virtual box 213 at a second time by applying the first velocity 217 to the first virtual box 211 identified at the first time. For example, the processor may identify the second position of the second virtual box 213 at the second time by applying the first velocity 217 to the first position of the first virtual box 211. The processor may determine the predicted displacement 215 based on a difference between the first location of the first virtual box 211 and the second location of the second virtual box 213.

[0072] With regard to a second example 203 in Fig. 2, the processor may identify a first virtual box 221 at the first time. The processor may identify a second virtual box 223 based on a plurality of points detected by a LiDAR (e.g., the LiDAR 120 in Fig.1) were obtained at a second time point.

[0073] For example, the processor may identify a first reference point 231 of the first virtual box 221 identified at the first time. For example, the processor may identify a second reference point 233 of the second virtual box 223 identified at the second time. The processor may determine a distance between the first reference point 231 and the second reference point 233. The processor may determine a first displacement 225 based on the distance between the first reference point 231 and the second reference point 233.

[0074] With reference to the first example 201 and the second example 203, the processor may determine a difference 227 between the predicted displacement 215 and the first displacement 225. The processor may determine an absolute value of the difference 227 between the predicted displacement 215 and the first displacement 225. The processor may determine whether the first velocity 217 is stable based on whether the absolute value of the difference 227 between the predicted displacement 215 and the first displacement 225 exceeds a threshold displacement. For example, the Fig. 2 described threshold shift in Fig. 1 described first to third threshold shifts.

[0075] As described above, the processor of the vehicle control device may determine whether the first speed is stable based on whether the difference between the first displacement 225 and the predicted displacement 215 exceeds the threshold displacement, and may select a speed to be assigned to the second virtual box based on whether the first speed is stable, thereby improving a vehicle control system with respect to the vehicle control device.

[0076] Fig. Figure 3 shows an example of identifying representative points of a virtual box.

[0077] With reference to Fig. 3, a processor (e.g., processor 110 in Fig. 1) a vehicle control device (e.g. the vehicle control device 100 of Fig.1) Identify representative points of virtual boxes 310 and 331. For example, virtual boxes 310 and 331 may include at least a first virtual box identified at a first time, a second virtual box identified at a second time, or any combination thereof.

[0078] With reference to a first example 301 in Fig. 3, the processor of the vehicle control device may identify the virtual box 310 corresponding to an external vehicle located in front of a vehicle 300. The processor may identify representative points 311, 313, 315, 317, 319, and 321 included in the virtual box 310.

[0079] For example, the processor may identify a maximum value of a second axis direction of points in a virtual box. The processor may identify a first vertex corresponding to the maximum value of the second axis direction as a representative point 311. For example, the processor may sequentially identify vertices (e.g., identified clockwise from the first vertex) based on identifying the first vertex corresponding to the maximum value of the second axis direction as the first representative point 311.

[0080] For example, the processor may identify a second vertex that can be identified first in a clockwise direction of the first vertex as the second representative point 313. The processor may identify a third vertex that can be identified first in a clockwise direction of the second vertex as the third representative point 315. The processor may identify a fourth vertex that can be identified first in a clockwise direction of the third vertex as the fourth representative point 317.

[0081] For example, if the virtual box 310 includes a line segment parallel to at least one of the first axis or the second axis, or any combination thereof, the processor may identify an upper-left vertex as the first representative point. The processor may sequentially identify vertices clockwise from the upper-left vertex as representative points based on the identification of the upper-left vertex as the first representative point.

[0082] The processor may identify the fifth representative point 319 between the first representative point 311 and the fourth representative point 317. The processor may identify the sixth representative point 321 between the second representative point 313 and the third representative point 315. For example, the processor may identify a midpoint between the first representative point 311 and the fourth representative point 317 as the fifth representative point 319. The processor may identify a midpoint between the second representative point 313 and the third representative point 315 as the sixth representative point 321.

[0083] For example, the processor may assign a first identifier to the first representative point 311. The processor may assign a second identifier to the second representative point 313. The processor may assign a third identifier to the third representative point 315. The processor may assign a fourth identifier to the fourth representative point 317. The processor may assign a fifth identifier to the fifth representative point 319. The processor may assign a sixth identifier to the sixth representative point 321.

[0084] For example, the first identifier may include an identifier indicating a top side. The second identifier may include an identifier indicating a right side. The third identifier may include an identifier indicating a bottom side. The fourth identifier may include an identifier indicating a left side. The fifth identifier may include an identifier indicating a top center. The sixth identifier may include an identifier indicating a bottom center.

[0085] The processor may identify the reference point 323 of the virtual box 310. The reference point 323 may include, for example, a reference point for at least one of a speed of an external vehicle corresponding to the virtual box 310, a location of the virtual box 310, or any combination thereof.

[0086] With reference to the second example 303 in Fig. 3, the processor of the vehicle control device may identify a virtual box 331 corresponding to an external vehicle identified before the vehicle 300.

[0087] For example, the processor may identify representative points 337, 339, 341, and 343 among a plurality of points 333 contained within the virtual box 331. For example, the processor may identify the representative points 337, 339, 341, and 343 among the plurality of points 333 contained within the virtual box 331 on a plane formed by a first axis and a second axis.

[0088] For example, the processor may identify the coordinates of each of the plurality of points 333 in the virtual box 331. The processor may identify the minimum value of the second axis from the plurality of points 333 contained in the virtual box 331. The processor may identify the seventh representative point 343 containing the minimum value of the second axis from the plurality of points 333 contained in the virtual box 331.

[0089] For example, the processor may identify the maximum value of the second axis from the plurality of points 333 contained in the virtual box 331. The processor may identify the eighth representative point 339 containing the maximum value of the second axis from the plurality of points 333 contained in the virtual box 331.

[0090] For example, the processor may identify the maximum value of the first axis from the plurality of points 333 contained in the virtual box 331. The processor may identify the ninth representative point 337 containing the maximum value of the first axis from the plurality of points 333 contained in the virtual box 331.

[0091] For example, the processor may identify the minimum value of the first axis from the plurality of points 333 contained in the virtual box 331. The processor may identify the tenth representative point 341 that contains the minimum value of the first axis from the plurality of points 333 in the virtual box 331.

[0092] The virtual box 331 in the second example 303 can be generated based on the multiple points 333 determined at the second time. A plurality of points 335 in the second example 303 can be determined at the first time.

[0093] The processor may apply an algorithm to the plurality of points 335 obtained at the first time and the plurality of points 333 obtained at the second time. For example, the processor may apply the algorithm to a first point cloud containing the plurality of points 335 obtained at the first time and a second point cloud containing the plurality of points 333 obtained at the second time. The algorithm described above may, for example, include an ICP algorithm.

[0094] The processor can Fig.1 is obtained based on the application of the algorithm to the plurality of points 335 obtained at the first time point and the plurality of points 333 obtained at the second time point, or to the first point cloud containing the plurality of points 335 obtained at the first time point and the second point cloud containing the plurality of points 333 obtained at the second time point.

[0095] Fig. Figure 4 shows an example in which a virtual box is identified in a boundary area of ​​a LiDAR.

[0096] With reference to Fig. 4, a processor (e.g., processor 110 in Fig. 1) a vehicle control device (e.g. the vehicle control device 100 of Fig. 1) virtual boxes 421, 423, 431 and 433 in border regions 411 and 413 of a LiDAR (e.g. the LiDAR 120 in Fig. 1) identify.

[0097] The processor may identify a direction in which the virtual box is located relative to the vehicle 400. The processor may identify some of the representative points based on the direction in which the virtual box is located relative to a vehicle 400.

[0098] The processor may identify the virtual box 423 in the first boundary region 411 corresponding to a left boundary region of the LiDAR. The processor may determine that the virtual box 423 is located in the first boundary region 411 of the LiDAR based on the virtual box 421 identified prior to identifying the virtual box 423.

[0099] The processor may identify the virtual box 433 in the second boundary region 413 corresponding to a right boundary region of the LiDAR. The processor may determine that the virtual box 433 is located in the second boundary region 413 of the LiDAR based on the virtual box 431 identified prior to identifying the virtual box 433.

[0100] The processor may identify representative points of the virtual box 423 that are identified in a left direction of the vehicle 400 and that are located in the first boundary region 411 of the LiDAR. For example, the processor may identify representative points of the virtual box 423 based on the identification of the virtual box 423 that are identified in the left direction of the vehicle 400 and that are located within a certain distance from the first boundary region 411 of the LiDAR. For example, the representative points of the virtual box 423 identified in the first boundary region 411 of the LiDAR may include some of the Fig. 3 identified representative points.

[0101] For example, the processor may identify a representative point indicating a top side of the virtual box 423, a representative point indicating a top center of the virtual box 423, a representative point indicating a maximum value of the first axis among a plurality of points included in the virtual box 423, a representative point indicating a minimum value of the second axis among the plurality of points included in the virtual box 423, and a representative point indicating a right side of the virtual box 423.

[0102] The processor may identify representative points of the virtual box 433 that are identified in a right direction of the vehicle 400 and that are located in the second boundary region 413 of the LiDAR. For example, the processor may identify representative points of the virtual box 433 based on the identification of the virtual box 433 that are identified in the right direction of the vehicle 400 and that are within a certain distance from the second boundary region 413 of the LiDAR. For example, the representative points of the virtual box 433 identified in the second boundary region 413 of the LiDAR may include some of the Fig. 3 identified representative points.

[0103] For example, the processor may identify a representative point indicating a top side of the virtual box 433, a representative point indicating a top center of the virtual box 433, a representative point indicating a maximum value of the first axis among a plurality of points included in the virtual box 433, a representative point indicating a maximum value of the second axis direction among the plurality of points included in the virtual box 433, and a representative point indicating a left side of the virtual box 433.

[0104] When the virtual box is identified in the boundary regions 411 and 413 of the LiDAR, the processor may obtain at least one of predicted displacement, or a second displacement, or any combination thereof by using some of the representative points.

[0105] As described above, when a virtual box is identified in the boundary region of the LiDAR, the processor of the vehicle control device may obtain at least one of the second displacement or the predicted displacement, or any combination thereof, by using some of the representative points to increase the reliability of a displacement.

[0106] Fig. 5 shows an example of determining a displacement based on first representative points of a virtual box identified at a first time and second representative points of a virtual box identified at a second time.

[0107] With reference to Fig. 5, a processor (e.g., processor 110 in Fig. 1) a vehicle control device (e.g. the vehicle control device 100 of Fig.1) Identify a predicted displacement 551. For example, the processor may determine a first speed of an external vehicle at a first time. The processor may apply the first speed to a first virtual box identified at the first time. The processor may obtain the predicted displacement 551 based on applying the first speed to the first virtual box identified at the first time.

[0108] The processor may identify a first virtual box 501 based on a plurality of points identified at the first time. The processor may identify a second virtual box 503 based on a plurality of points identified at the second time.

[0109] For example, the processor may identify first representative points included in the first virtual box 501. The first representative points in the first virtual box 501 may, for example, include at least one of the Fig. 3 representative points described.

[0110] For example, the processor may identify second representative points included in the second virtual box 503. The second representative points in the second virtual box 503 may, for example, include at least one of the Fig. 3 representative points described.

[0111] The processor may determine the displacements 511, 513, 521, 523, 531, 533, and 541 based on the first representative points contained in the first virtual box 501 and the second representative points contained in the second virtual box 503. For example, the processor may identify the displacements 511, 513, 521, 523, 531, 533, and 541 based on a difference between the first representative points contained in the first virtual box 501 and the second representative points corresponding to the first representative points and contained in the second virtual box 503.

[0112] The processor may select a second displacement 561 from the displacements 511, 513, 521, 523, 531, 533 and 541 that has the smallest difference from the predicted displacement 551.

[0113] In one embodiment, the processor may determine a second velocity, a third velocity, or a fourth velocity using the selected second displacement 561.

[0114] For example, the processor may determine the second velocity by dividing the second displacement 561 by a time difference between the first time at which the first virtual box 501 was obtained and the second time at which the second virtual box 503 was obtained.

[0115] For example, the processor may determine a third displacement based on applying the algorithm to the first points in the first virtual box 501 and the second points in the second virtual box 503. The processor may determine an average of the third displacement and the second displacement 561. The processor may determine the third velocity based on dividing the average of the third displacement and the second displacement 561 by the time difference between the first time the first virtual box 501 was obtained and the second time the second virtual box 503 was obtained.

[0116] For example, the processor may determine a fourth displacement based on the second displacement 561 and the predicted displacement 551. The processor may apply a weight to each of the second displacement 561 and the predicted displacement 551 based on a difference between the second displacement 561 and the predicted displacement 551 that is greater than or equal to a threshold. For example, the processor may apply a first weight to the second displacement 561 and a second weight to the predicted displacement 551. The processor may determine the fourth displacement based on dividing the sum of the second displacement 561 with the first weight applied and the predicted displacement 551 with the second weight applied by 2.The processor may obtain the fourth velocity by dividing by 2 the fourth displacement, which is the sum of the second displacement 561 to which the first weight is applied and the predicted displacement 551 to which the second weight is applied, by the time difference between the first time at which the first virtual box 501 was determined and the second time at which the second virtual box 503 was determined.

[0117] The processor may assign one of the obtained second, third, or fourth speeds to the second virtual box 503.

[0118] As described above, the processor of the vehicle control device can stabilize the speed of the second virtual box 503 by assigning one of the second speed, the third speed, or the fourth speed to the second virtual box 503. The processor can reduce an error in the vehicle control system connected to the vehicle control device by assigning one of the second speed, the third speed, or the fourth speed to the second virtual box 503. The vehicle control system may include, for example, a control system that assists driving the vehicle.

[0119] Fig. 6 shows an example of a result to which one or more aspects of the present disclosure are applied and an example of a result to which the one or more aspects of the present disclosure are not applied.

[0120] A first example 601 in Fig. 6 may include an example expressing a longitudinal speed of a vehicle. A second example 603 in Fig. 6 may include an example expressing a lateral velocity of the vehicle.

[0121] When the present technology is not applied, outliers 611 and 621 may be identified in a first diagram of the first example 601 and in a second diagram of the second example 603, respectively. The cases where the outliers 611 and 621 are identified may include cases where the speed of an external vehicle suddenly changes. When the speed of the external vehicle suddenly changes, an error may occur while the vehicle control system connected to the vehicle control device is performing vehicle driving assistance.

[0122] Accordingly, the vehicle control device (e.g., the vehicle control device 100 of Fig. 1) Reduce the occurrence of outliers 611 and 621 by applying the present technology. For example, the vehicle control device can stably output at least one of a longitudinal speed of the external vehicle and a lateral speed of the external vehicle, or any combination thereof, by performing various operations depending on the speed of the external vehicle.

[0123] The vehicle control system can reduce the root mean square error (RMSE) by applying the present technology. The vehicle control system can obtain ground truth (GT) depending on the driving situation and can reduce errors in the speed of external objects by applying the present technology.

[0124] Hereinafter, a vehicle control method according to the present disclosure will be described with reference to Fig. 7 described in detail. Fig. 7 shows an example of a flowchart for a vehicle control method.

[0125] In the following, it is assumed that the vehicle control device 100 of Fig. 1 the process of Fig. 7. In addition, when describing Fig. 7 that an operation described as being performed by a processor is controlled by the processor 110 of the vehicle control device 100.

[0126] At least one of the processes in Fig. 7 can be controlled by the vehicle control device 100 from Fig. 1. Each of the operations in Fig. 7 can be executed sequentially, but they don't necessarily have to be executed sequentially. For example, the order of operations can be changed, and at least two operations can be executed in parallel.

[0127] Referring to Fig. 7, a vehicle control method may, in operation S701, include a process of identifying a first speed of an external vehicle at the first time point based on the identification of a first virtual box corresponding to the external vehicle at the first time point by a LiDAR (e.g., the LiDAR 120 in Fig. 1) include.

[0128] In operation S703, the vehicle control method may include an operation of determining whether a confidence level of the first speed exceeds a reference confidence level based on at least one of the following information: information of the first virtual box, information of a second virtual box identified at a second time point after the first time point, or driving information of a vehicle, or any combination thereof.

[0129] The vehicle control method may include an operation for identifying the confidence level of the first speed based on at least one of the following: a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the vehicle identified at the first time and a second yaw rate of the vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are in a boundary area of ​​the LiDAR or whether the first virtual box or the second virtual box is occluded, or any combination thereof.

[0130] For example, the boundary region of the LiDAR may include a region within a certain distance from a boundary line of the LiDAR's field of view (FOV).

[0131] The vehicle control method may include an operation for increasing the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box is less than or equal to a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box is less than or equal to a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to a reference width, the first virtual box and the second virtual box are not located in the boundary region of the LiDAR, a component of a first axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than a component of a second axis direction,the first virtual box and the second virtual box are not obscured, or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time is less than or equal to a reference yaw rate.

[0132] The vehicle control method may include a process of initializing the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box exceeds a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box exceeds a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box exceeds a reference width, the difference between a speed of the vehicle at the first time and the speed of the vehicle at the second time exceeds a reference speed,or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time exceeds a reference yaw rate.

[0133] In operation S705, the vehicle control method may include an operation of determining whether the first speed is stable based on a first displacement obtained by a difference between a first position of the first virtual box and a second position of the second virtual box, a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time point, and a predetermined threshold displacement when the confidence level of the first speed exceeds the reference confidence level.

[0134] The vehicle control method may include a process of determining that the first speed is unstable based on a difference between the first displacement and the predicted displacement that exceeds the threshold displacement. The vehicle control method may include a process of determining that the first speed is stable based on the difference between the first displacement and the predicted displacement that is less than or equal to the threshold displacement.

[0135] The vehicle control method may include an operation of assigning the first speed to the second virtual box when the confidence level of the first speed is less than or equal to a first reference confidence level.

[0136] The vehicle control method may include determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a first threshold displacement when the confidence level of the first speed exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level.

[0137] The vehicle control method may include determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is smaller than the first threshold displacement when the confidence level of the first speed exceeds the second reference confidence level.

[0138] The vehicle control method may include an operation of determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement that exceeds the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary area of ​​the LiDAR.

[0139] In operation S707, the vehicle control method may include a process of selecting a second displacement having the smallest difference from the predicted displacement from among displacements based on a difference between first representative points included in the first virtual box and second representative points each corresponding to the first representative points and included in the second virtual box when the first speed is determined to be unstable.

[0140] The vehicle control method may include an operation of identifying, as at least one of the first representative points and the second representative points, at least one of a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, from the vertices of the first virtual box and the second virtual box, and points included in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis.

[0141] The vehicle control method may include an operation of determining some of the first representative points or some of the second representative points based on a direction in which the first virtual box is located relative to the vehicle when the first virtual box or the second virtual box is identified in a boundary area of ​​the LiDAR.

[0142] In operation S709, the vehicle control method may include a process of assigning to the second virtual box one of a second speed based on the selected second displacement, a third speed based on the second displacement and a third displacement obtained by applying an algorithm to first points in the first virtual box and second points in the second virtual box, or a fourth speed based on the second displacement and the predicted displacement.

[0143] The vehicle control method may include a process of assigning the second speed to the second virtual box when the first virtual box and the second virtual box are identified as exceeding a designated distance from the vehicle so as not to apply the algorithm to the first points and the second points.

[0144] The vehicle control method may include a process of determining the third displacement based on applying the algorithm to the first points and the second points when the first virtual box and the second virtual box are identified at a designated distance or less from the vehicle to apply the algorithm to the first points and the second points. The vehicle control method may include a process of assigning the third speed based on the average of the third displacement and the second displacement to the second virtual box.

[0145] The vehicle control method may include a process of obtaining a fourth displacement by applying a weight to the second displacement and the predicted displacement based on a difference between the second displacement and the predicted displacement that is greater than or equal to a threshold. According to one embodiment, the vehicle control method may include a process of assigning the fourth speed to the second virtual box based on the determination of the fourth speed using the fourth displacement.

[0146] As described above, the vehicle control method according to an embodiment may perform various operations to stabilize the speed of the second virtual box identified at the second time point when the first speed of the external vehicle identified at the first time point is not stable.According to one embodiment, the vehicle control method may assign to the second virtual box one of a second speed based on the selected second displacement, a third speed based on the second displacement and the third displacement obtained by applying the algorithm to the first points in the first virtual box and the second points in the second virtual box, or a fourth speed based on the second displacement and the predicted displacement, thereby stabilizing a speed of the second virtual box identified at the second time point.

[0147] Hereinafter, a vehicle control method according to an embodiment of the present disclosure will be described with reference to Fig. 8 described in detail. Fig. 8 shows an example of a flowchart relating to a vehicle control method.

[0148] In the following, it is assumed that the vehicle control device 100 of Fig. 1 the process of Fig. 8. In addition, the description of Fig. 8, it can be assumed that a process described as being performed by a processor is controlled by the processor 110 of the vehicle control device 100.

[0149] At least one of the processes in Fig. 8 can be controlled by the vehicle control device 100 from Fig. 1. Each of the operations in Fig. 8 can be executed sequentially, but they don't necessarily have to be executed sequentially. For example, the order of operations can be changed, and at least two operations can be executed in parallel.

[0150] Referring to Fig. 8, a vehicle control method may include, in operation S801, a process of determining whether a first speed of an external vehicle identified at a first time exceeds a reference confidence level.

[0151] If the first speed of the external vehicle identified at the first time point does not exceed the reference confidence level (No in operation S801), the vehicle control method may include an operation of determining, in operation S803, whether at least one of a first virtual box corresponding to the external vehicle identified at the first time point or a second virtual box corresponding to the external vehicle identified at a second time point after the first time point is identified in a boundary region of a field of view (FOV) of a LiDAR.

[0152] If the first speed of the external vehicle identified at the first time exceeds the reference confidence level (Yes in operation S801), or at least one of the first virtual box or the second virtual box is identified in the FOV boundary region (Yes in operation S803), the vehicle control method may include, in operation S805, a process of setting a threshold value depending on a confidence level of the first speed. The threshold value may, for example, be the Fig. 1 described threshold shift.

[0153] In operation S807, the vehicle control method may include a process of calculating a predicted displacement based on the speed of the external vehicle. For example, the vehicle control method may include a process of calculating a predicted displacement corresponding to the changes in the first virtual box and the second virtual box position using the first speed of the external vehicle.

[0154] For example, the vehicle control method may include an operation of estimating a position of the second virtual box by applying the first speed of the external vehicle to the first virtual box and calculating the predicted displacement based on a difference between the estimated position of the second virtual box and the position of the first virtual box.

[0155] In operation S809, the vehicle control method may include a process of determining whether the speed of the external vehicle is stable. For example, the vehicle control method may include a process of determining whether the first speed of the external vehicle is stable. For example, the vehicle control method may include a process of determining whether the first speed is stable based on a difference between the predicted displacement obtained in operation S807 and an observed displacement obtained based on the difference between the observed position of the first virtual box and the observed position of the second virtual box.

[0156] When the speed of the external vehicle is unstable (No in operation S809), the vehicle control method may include an operation for analyzing the displacement of each representative point in operation S811.

[0157] For example, the vehicle control method may include a process of identifying first representative points of the first virtual box determined at the first time. For example, the vehicle control method may include a process of identifying second representative points of the second virtual box determined at the second time. The vehicle control method may include a process of identifying the second representative points corresponding to the first representative points, respectively, and comparing the locations of the first representative points and the locations of the second representative points corresponding to the first representative points.The vehicle control method may include a process for determining displacements of the first representative points and the second representative points each corresponding to the first representative points based on the comparison of the locations of the first representative points with the locations of the second representative points each corresponding to the first representative points.

[0158] In operation S813, the vehicle control method may include a process of comparing a displacement of each representative point with a displacement based on the speed of the external vehicle. For example, the displacement based on the speed of an external vehicle may include a predicted displacement.

[0159] In operation S815, the vehicle control method may include an operation of determining whether a displacement of each of the representative points satisfies a condition based on comparing the displacement of each of the representative points with the predicted displacement.

[0160] For example, the vehicle control method may include a process of comparing the displacement of each of the representative points with the predicted displacement. The vehicle control method may include a process of determining a displacement with the smallest difference from the predicted displacement based on the comparison of the displacement of each of the representative points with the predicted displacement.

[0161] If the condition is met (Yes in operation S815), the vehicle control method may include an operation for updating a new displacement in operation S817.

[0162] For example, the vehicle control method may include an operation for updating the displacement with the smallest deviation from the predicted displacement so that it is changed to a new displacement.

[0163] For example, the vehicle control method may include an operation for updating the displacement obtained by the ICP algorithm such that the new displacement is changed when the displacement obtained by an ICP algorithm has the smallest difference from the predicted displacement.

[0164] For example, the vehicle control method may include a process of determining that the displacement to be updated to the new displacement is unstable when a difference between the displacement to be updated to the new displacement and the predicted displacement is greater than or equal to a threshold.

[0165] The vehicle control method may include an operation in which a designated operation (e.g., a weight sum operation) is applied to the predicted displacement and the updated displacement to be changed to the new displacement when the updated displacement to be changed to the new displacement is unstable.

[0166] In operation S819, the vehicle control method may include a process of using the updated displacement. For example, the vehicle control method may include a process of calculating a speed to be assigned to the second virtual box using the updated displacement. For example, the vehicle control method may include a process of assigning the calculated speed to the second virtual box based on the calculation of the speed to be assigned to the second virtual box using the updated displacement.

[0167] When the first virtual box or the second virtual box is not identified in a FOV area (No in operation S803), the speed of the external vehicle is stable (Yes in operation S809), or the condition is not satisfied (No in operation S815), the vehicle control method may include a process of using a current displacement in operation S821.

[0168] For example, the vehicle control method may include a process for determining the current displacement based on the observed position of the first virtual box and the observed position of the second virtual box. For example, the vehicle control method may include a process for calculating a speed assigned to the second virtual box based on the current displacement.

[0169] According to one aspect of the present disclosure, a vehicle control device may include a light detection and ranging device (LiDAR) and a processor. The processor may identify a first speed of an external vehicle at a first time based on the LiDAR's identification of a first virtual box corresponding to the external vehicle at the first time, may determine whether a confidence level of the first speed exceeds a reference confidence level based on at least one of information of the first virtual box, information of a second virtual box identified at a second time after the first time, or driving information of a vehicle, or any combination thereof, may be based on a first displacement,which is obtained by a difference between a first position of the first virtual box and a second position of the second virtual box, a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time, and a predetermined threshold displacement, determine whether the first speed is stable when the confidence level of the first speed exceeds the reference confidence level, may select a second displacement having the smallest difference from the predicted displacement from displacements based on a difference between first representative points included in the first virtual box and second representative points each corresponding to the first representative points and included in the second virtual box when the first speed is determined to be unstable,and may assign to the second virtual box one of a second velocity based on the selected second displacement, a third velocity based on the second displacement and a third displacement obtained by applying an algorithm to first points contained in the first virtual box and second points contained in the second virtual box, or a fourth velocity based on the second displacement and the predicted displacement.

[0170] In one example, the processor may identify the confidence level of the first speed based on at least one of a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the vehicle identified at the first time and a second yaw rate of the vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are located in a boundary region of the LiDAR, or whether the first virtual box or the second virtual box is occluded, or any combination,and wherein the boundary region of the LiDAR comprises a region within a specified distance from a boundary line of a field of view (FOV) of the LiDAR. The boundary region of the LiDAR may comprise a region within a designated distance from a boundary line of a field of view (FOV) of the LiDAR.

[0171] In one example, the processor may increase the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box is less than or equal to a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box is less than or equal to a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to a reference width, the first virtual box and the second virtual box are not located in the boundary region of the LiDAR, a component of a first axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than a component of a second axis direction,the first virtual box and the second virtual box are not obscured, or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time is less than or equal to a reference yaw rate.

[0172] In one example, the processor may initialize the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box exceeds a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box exceeds a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box exceeds a reference width, the difference between a speed of the vehicle at the first time and the speed of the vehicle at the second time exceeds a reference speed,or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time exceeds a reference yaw rate.

[0173] In one example, the processor may determine that the first speed is unstable based on a difference between the first displacement and the predicted displacement that exceeds the threshold, and may determine that the first speed is stable based on the difference between the first displacement and the predicted displacement that is less than or equal to the threshold.

[0174] In one example, the processor may assign the first velocity to the second virtual box if the confidence level of the first velocity is less than or equal to a first reference confidence level, may determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement that exceeds a first threshold displacement, if the confidence level of the first velocity exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level, may determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement that exceeds a second threshold displacement that is less than the first threshold displacement,when the confidence level of the first velocity exceeds the second reference confidence level, and may determine that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement, which exceeds the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary region of the LiDAR. The boundary region of the LiDAR may include a region within a certain distance from a boundary line of a FOV of the LiDAR.

[0175] In one example, the processor may identify as at least one of the first representative points and the second representative points at least one of the vertices of the first virtual box and the second virtual box, a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, and points included in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis.

[0176] In one example, the processor may determine some of the first representative points or some of the second representative points based on a direction in which the first virtual box is located relative to the vehicle when the first virtual box or the second virtual box is identified within a boundary region of the LiDAR. The boundary region of the LiDAR may include a region within a certain distance from a boundary line of a field of view of the LiDAR.

[0177] In one example, if the first virtual box and the second virtual box are identified as exceeding a certain distance from the vehicle, the processor may assign the second speed to the second virtual box so as not to apply the algorithm to the first points and the second points.

[0178] In one example, the processor may determine the third displacement based on applying the algorithm to the first points and the second points when the first virtual box and the second virtual box are identified at a designated distance or less from the vehicle to apply the algorithm to the first points and the second points, and may assign the third speed based on an average of the third displacement and the second displacement to the second virtual box.

[0179] In one example, the processor may determine a fourth displacement by applying a weight to each of the second displacement and the predicted displacement based on a difference between the second displacement and the predicted displacement that is greater than or equal to a threshold, and may assign the fourth velocity to the second virtual box based on obtaining the fourth velocity using the fourth displacement.

[0180] According to one aspect of the present disclosure, a vehicle control method may include identifying a first speed of an external vehicle at a first time point based on identifying a first virtual box corresponding to the external vehicle at the first time point by a LiDAR, determining whether a confidence level of the first speed exceeds a reference confidence level based on at least one of information of the first virtual box, information of a second virtual box identified at a second time point after the first time point, driving information of a vehicle, or any combination thereof, determining whether the first speed is stable based on a first displacement obtained by a difference between a first location of the first virtual box and a second location of the second virtual box,a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time point and a predetermined threshold displacement when the confidence level of the first speed exceeds the reference confidence level; selecting a second displacement having the smallest difference from the predicted displacement from among the displacements based on a difference between first representative points included in the first virtual box and second representative points each corresponding to the first representative points and included in the second virtual box when the first speed is determined to be unstable; and assigning a second speed based on the selected second displacement, a third speed based on the second displacement, and a third displacement;obtained by applying an algorithm to first points contained in the first virtual box and second points contained in the second virtual box, or a fourth velocity based on the second displacement and the predicted displacement to the second virtual box.

[0181] According to one example, the vehicle control method may further include identifying the confidence level of the first speed based on at least one of the following: a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the vehicle identified at the first time and a second yaw rate of the vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are in a boundary region of the LiDAR or whether the first virtual box or the second virtual box is occluded, or any combination thereof.The boundary region of the LiDAR may include a region within a certain distance from a boundary line of a field of view (FOV) of the LiDAR.

[0182] According to one example, the vehicle control method may further include increasing the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box is less than or equal to a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box is less than or equal to a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to a reference width, the first virtual box and the second virtual box are not located in the boundary region of the LiDAR, a component of a first axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box,is greater than a component of a second axis direction, the first virtual box and the second virtual box are not obscured, or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time is less than or equal to a reference yaw rate.

[0183] According to one example, the vehicle control method may further comprise initializing the confidence level of the first speed based on the fact that the difference between the first size of the first virtual box and the second size of the second virtual box exceeds a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box exceeds a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box exceeds a reference width, the difference between a speed of the vehicle at the first time and the speed of the vehicle at the second time exceeds a reference speed,or the difference between the first yaw rate of the vehicle identified at the first time and the second yaw rate of the vehicle identified at the second time exceeds a reference yaw rate.

[0184] According to one example, the vehicle control method may further include determining that the first speed is unstable based on a difference between the first displacement and the predicted displacement exceeding the threshold, and determining that the first speed is stable based on the difference between the first displacement and the predicted displacement being less than or equal to the threshold.

[0185] According to one example, the vehicle control method may further comprise: assigning the first speed to the second virtual box if the confidence level of the first speed is less than or equal to a first reference confidence level; determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a first threshold displacement; if the confidence level of the first speed exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level; determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is less than the first threshold displacement;if the confidence level of the first velocity exceeds the second reference confidence level, and determining that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement that exceeds the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary region of the LiDAR. The boundary region of the LiDAR may include a region within a certain distance from a boundary line of a FOV of the LiDAR.

[0186] According to an example, the vehicle control method may further include identifying, as at least one of the first representative points and the second representative points, at least one of the vertices of the first virtual box and the second virtual box, a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, and points included in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis.

[0187] According to one example, the vehicle control method may further include obtaining some of the first representative points or some of the second representative points based on a direction in which the first virtual box is located relative to the vehicle when the first virtual box or the second virtual box is identified in a boundary region of the LiDAR. The boundary region of the LiDAR may include a region within a certain distance from a boundary line of a field of view of the LiDAR.

[0188] According to one example, the vehicle control method may further include assigning the second speed to the second virtual box when the first virtual box and the second virtual box are identified as exceeding a designated distance from the vehicle so as not to apply the algorithm to the first points and the second points, obtaining the third displacement based on applying the algorithm to the first points and the second points when the first virtual box and the second virtual box are identified at a designated distance or less from the vehicle so as to apply the algorithm to the first points and the second points, assigning the third speed based on an average of the third displacement and the second displacement to the second virtual box,Obtaining a fourth displacement by applying a weight to each of the second displacement and the predicted displacement based on a difference between the second displacement and the predicted displacement that is greater than or equal to a threshold, and assigning the fourth velocity to the second virtual box based on obtaining the fourth velocity by using the fourth displacement.

[0189] Fig. 9 shows an example of a computer system associated with a vehicle control device or method.

[0190] Referring to Fig. 9, a computer system 1000 may include at least a processor 1100, a memory 1300, a user interface input device 1400, a user interface output device 1500, a mass storage 1600, and a network interface 1700 interconnected via a bus 1200.

[0191] Processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in memory 1300 and / or mass storage 1600. Both memory 1300 and mass storage 1600 may include various types of volatile or non-volatile storage media. For example, memory 1300 may include read-only memory (ROM) and random access memory (RAM).

[0192] Accordingly, the operations of the method or algorithm described in connection with the embodiment(s) disclosed in the specification may be implemented directly with a hardware module or a software module, or any combination thereof, executed by processor 1100. The software module may be located on a storage medium (i.e., memory 1300 and / or mass storage 1600), such as random access memory (RAM), flash memory, read-only memory (ROM), erasable programmable read-only memory (EPROM), electrical EPROM (EEPROM), a register, a hard disk drive, a removable disk, or a compact disc-ROM (CD-ROM).

[0193] The storage medium may be connected to the processor 1100. The processor 1100 may read information from the storage medium and write information to the storage medium. Alternatively, the storage medium may be integrated into the processor 1100. The processor and the storage medium may be implemented with an application-specific integrated circuit (ASIC). The ASIC may be housed in a user terminal. Alternatively, the processor and the storage medium may be implemented with separate components in the user terminal.

[0194] The above description is merely an example of the technical idea of ​​the present disclosure, and various modifications and alterations can be made by a person skilled in the art without departing from the essential features of the present disclosure.

[0195] Accordingly, various examples of the present disclosure are intended to illustrate the technical idea of ​​the present disclosure, not to limit it, and the scope and spirit of the present disclosure are not limited by the above examples. The scope of the present disclosure should be interpreted by the appended claims, and all equivalents thereof should be construed as being included within the scope of the present disclosure.

[0196] The present technology may identify a speed to be assigned to a virtual box using at least one of a speed of an external vehicle or a displacement of the virtual box corresponding to the external vehicle, or any combination thereof.

[0197] In addition, the present technology can detect a virtual box with an unstable speed and assign a stable speed to the virtual box with the unstable speed.

[0198] In addition, the present technology can reduce a speed error caused by a shape error of an object.

[0199] In addition, a variety of effects can be provided that are understood directly or indirectly through the specification.

[0200] Although the present disclosure has been described with reference to 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 to which the present disclosure pertains without departing from the spirit and scope of the present disclosure as claimed in the following claims.

Claims

[1] Device for a first vehicle, the device comprising: at least one sensor; and a processor, wherein the processor is configured to: to determine, on the basis of the at least one sensor, a first virtual box corresponding to an external vehicle and associated with a first point in time; to identify a first speed of the external vehicle at the first time point based on the first virtual box; determine whether a confidence level of the first speed exceeds a reference confidence level based on at least one of information of the first virtual box, information of a second virtual box identified at a second time after the first time, or driving information of the first vehicle, or any combination thereof; to determine whether the first speed is stable based on a first displacement determined by a difference between a first position of the first virtual box and a second position of the second virtual box, a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time point, and a predetermined threshold displacement after the confidence level of the first speed exceeds the reference confidence level; selecting a second displacement having the smallest difference from the predicted displacement from among the displacements based on a difference between first representative points contained in the first virtual box and second representative points, each corresponding to the first representative points and contained in the second virtual box, after the first velocity has been determined to be unstable; and the second virtual box a second speed based on the selected second displacement, a third speed based on the second displacement and a third displacement, obtained by applying an algorithm to first points in the first virtual box and second points in the second virtual box, or to assign a fourth velocity based on the second displacement and the predicted displacement. [2] The device of claim 1, wherein the processor is configured to: to identify the confidence level of the first speed based on at least one of a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the first vehicle identified at the first time and a second yaw rate of the first vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are in a boundary region of the at least one sensor, or whether the first virtual box or the second virtual box is occluded, or any combination thereof, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [3] The device of claim 2, wherein the processor is configured to: to increase the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box being less than or equal to a reference rate, the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box being less than or equal to a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box being less than or equal to a reference width, the first virtual box and the second virtual box are not located in the boundary region of the at least one sensor, a component of a first axis direction contained in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than a component of a second axis direction,the first virtual box and the second virtual box are not obscured or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time is less than or equal to a reference yaw rate. [4] The device of claim 2, wherein the processor is configured: initialize the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box exceeding a reference rate, the difference between the first heading of the first virtual box and the second heading of the second virtual box exceeding a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box exceeding a reference width, the difference between a speed of the first vehicle at the first time and the speed of the first vehicle at the second time exceeding a reference speed, or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time exceeding a reference yaw rate. [5] The device of claim 1, wherein the processor is configured to: Determine that the first velocity is unstable based on a difference between the first displacement and the predicted displacement exceeding the threshold displacement; and to determine that the first velocity is stable based on the difference between the first displacement and the predicted displacement being less than or equal to the threshold displacement. [6] The device of claim 5, wherein the processor is configured to: assign the first speed to the second virtual box based on the confidence level of the first speed being less than or equal to a first reference confidence level; determine that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a first threshold displacement when the confidence level of the first speed exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level; determine that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is smaller than the first threshold displacement when the confidence level of the first speed exceeds the second reference confidence level; and to determine that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement exceeding the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary region of the at least one sensor, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [7] The device of claim 1, wherein the processor is configured to: as at least one of the first representative points and the second representative points, at least one of the vertices of the first virtual box and the second virtual box, a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, and points contained in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis. [8] The device of claim 1, wherein the processor is configured to: to determine at least one of the first representative points or at least one of the second representative points based on a direction in which the first virtual box is located relative to the first vehicle, based on the first virtual box or the second virtual box being identified in a boundary area of the at least one sensor, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [9] The device of claim 1, wherein the processor is configured to: to assign the second speed to the second virtual box based on the first virtual box and the second virtual box being identified as exceeding a certain distance from the first vehicle, so as not to apply the algorithm to the first points and the second points. [10] The device of claim 9, wherein the processor is configured to: to determine the third displacement based on applying the algorithm to the first points and the second points, based on identifying the first virtual box and the second virtual box at a certain distance or less from the first vehicle to apply the algorithm to the first points and the second points; and assign the third velocity, based on an average of the third displacement and the second displacement, to the second virtual box. [11] The device of claim 1, wherein the processor is configured to: determine a fourth displacement by applying a weight to each of the second displacement and the predicted displacement based on a difference between the second displacement and the predicted displacement being greater than or equal to a threshold; and to assign the fourth velocity to the second virtual box based on obtaining the fourth velocity using the fourth displacement. [12] A method performed by a device of a first vehicle, the method comprising: Determining a first virtual box corresponding to an external vehicle and associated with a first time based on at least one sensor; Identifying a first speed of the external vehicle at the first time based on the first virtual box; Determining whether a confidence level of the first speed exceeds a reference confidence level based on at least one of information of the first virtual box, information of a second virtual box identified at a second time after the first time, or driving information of the first vehicle, or any combination thereof; Determining whether the first speed is stable based on a first displacement obtained by a difference between a first position of the first virtual box and a second position of the second virtual box, a predicted displacement of the first virtual box predicted by the first speed of the external vehicle at the first time point, and a predetermined threshold displacement after the confidence level of the first speed exceeds the reference confidence level; Selecting a second displacement having the smallest difference from the predicted displacement from displacements based on a difference between first representative points contained in the first virtual box and second representative points each corresponding to the first representative points and contained in the second virtual box after the first velocity has been determined to be unstable; and Assigning, to the second virtual box, one of a second velocity based on the selected second displacement, a third velocity based on the second displacement and a third displacement obtained by applying an algorithm to first points in the first virtual box and second points in the second virtual box, or a fourth velocity based on the second displacement and the predicted displacement. [13] The method of claim 12, further comprising: Identifying the confidence level of the first speed based on at least one of a difference between a first size of the first virtual box and a second size of the second virtual box, a difference between a first heading of the first virtual box and a second heading of the second virtual box, a difference between a first yaw rate of the first vehicle identified at the first time and a second yaw rate of the first vehicle identified at the second time, a difference between a first width of the first virtual box and a second width of the second virtual box, whether the first location and the second location are in a boundary area of the at least one sensor or whether the first virtual box or the second virtual box is occluded, or any combination thereof, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [14] The method of claim 13, further comprising: Increasing the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box being less than or equal to a reference rate, wherein the difference between the first heading direction of the first virtual box and the second heading direction of the second virtual box is less than or equal to a reference angle, wherein the difference between the first width of the first virtual box and the second width of the second virtual box is less than or equal to a reference width, the first virtual box and the second virtual box are not located in the boundary region of the at least one sensor, a component of a first axis direction included in the first heading direction of the first virtual box and the second heading direction of the second virtual box is greater than a component of a second axis direction,the first virtual box and the second virtual box are not obscured or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time is less than or equal to a reference yaw rate. [15] The method of claim 13, further comprising: Initializing the confidence level of the first speed based on the difference between the first size of the first virtual box and the second size of the second virtual box that exceeds a reference rate, the difference between the first heading of the first virtual box and the second heading of the second virtual box that exceeds a reference angle, the difference between the first width of the first virtual box and the second width of the second virtual box that exceeds a reference width, the difference between a speed of the first vehicle at the first time and the speed of the first vehicle at the second time that exceeds a reference speed, or the difference between the first yaw rate of the first vehicle identified at the first time and the second yaw rate of the first vehicle identified at the second time,which exceeds a reference yaw rate. [16] The method of claim 12, further comprising: Determine that the first velocity is unstable based on a difference between the first displacement and the predicted displacement that exceeds the threshold displacement. [17] The method of claim 16, further comprising: Assigning the first velocity to the second virtual box based on the confidence level of the first velocity being less than or equal to a first reference confidence level; Determining that the first velocity is unstable based on the difference between the first displacement and the predicted displacement exceeding a first threshold displacement when the confidence level of the first velocity exceeds the first reference confidence level and is less than or equal to a second reference confidence level that exceeds the first reference confidence level; Determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a second threshold displacement that is less than the first threshold displacement when the confidence level of the first speed exceeds the second reference confidence level; and Determining that the first speed is unstable based on the difference between the first displacement and the predicted displacement exceeding a third threshold displacement that exceeds the first threshold displacement when the first virtual box or the second virtual box is identified in a boundary region of the at least one sensor, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [18] The method of claim 12, further comprising: Identifying, as at least one of the first representative points and the second representative points, at least one of the vertices of the first virtual box and the second virtual box, a minimum value of a first axis, a maximum value of the first axis, a minimum value of a second axis, or a maximum value of the second axis, or any combination thereof, and points included in the first virtual box and the second virtual box on a plane formed by the first axis and the second axis under the first axis, the second axis, and a third axis. [19] The method of claim 12, further comprising: Determining at least one of the first representative points or at least one of the second representative points based on a direction in which the first virtual box is located relative to the first vehicle, based on the first virtual box or the second virtual box being identified in a boundary area of the at least one sensor, and wherein the boundary region of the at least one sensor comprises a region within a certain distance from a boundary line of a field of view (FOV) of the at least one sensor. [20] The method of claim 12, further comprising: Assigning the second speed to the second virtual box based on identifying the first virtual box and the second virtual box as exceeding a certain distance from the first vehicle so as not to apply the algorithm to the first points and the second points.

Images