Object recognition device and method thereof

Abstract

An object recognition apparatus and a method thereof are provided. The apparatus may include: a sensor; a memory configured to store a reference grid map representing an appearance of the vehicle; and a processor configured to project a plurality of points obtained by the sensor representing the external object onto the first surface to obtain a first grid map including a first plurality of grids, and to project a plurality of points indicative of the external object onto the second surface to obtain a second grid map including a second plurality of grids, and determining a type of the external object based on at least one of the reference grid maps and at least one of the first grid map or the second grid map, and outputting a signal indicating the type of the external object.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of priority from Korean Patent Application No. 10-2023-0124149 filed in the Korean Intellectual Property Office on September 18, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to an object recognition device and method thereof, and more particularly, to a technology for recognizing a type of an external object using a sensor, such as a light detection and ranging (LiDAR) sensor. Background Art

[0004] Various studies are underway to use various sensors to recognize external objects to assist in driving a host vehicle.

[0005] Specifically, when the host vehicle is operating in a driving assistance device activation mode or an autonomous driving mode, external objects may be identified using sensors (eg, LiDAR sensors).

[0006] If the external object is separated and identified, the external object is incorrectly identified using LiDAR, thereby causing an error in the vehicle control system. If a single external object is separated and identified, it may be identified as a static state although the external object is operating, causing a problem that the driving route of the host vehicle changes rapidly. Summary of the invention

[0007] According to the present disclosure, a device may include: a sensor; a memory configured to store a reference grid map representing the appearance of a vehicle; and a processor configured to project multiple points indicating an external object onto a first surface to obtain a first grid map including a first plurality of grids, wherein the external object is obtained based on sensing information obtained by the sensor, and the first surface is formed by a second axis and a third axis among a first axis, a second axis, and a third axis, wherein the first axis is parallel to a driving direction of the vehicle and perpendicular to the second axis, and the third axis is perpendicular to the first axis and the second axis, project multiple points indicating the external object onto a second surface to obtain a second grid map including a second plurality of grids, wherein the second surface is formed by the first axis and the third axis, determine the type of the external object based on at least one of the reference grid maps and at least one of the first grid map and the second grid map, and output a signal indicating the type of the external object.

[0008] In the apparatus, wherein the processor is configured to display a map including the external object based on the signal indicating the type of the external object and perform autonomous driving control of the vehicle based on the type of the external object.

[0009] In the device, the processor is configured to acquire at least one of the first grid map or the second grid map based on the external object being determined to be in a moving state.

[0010] In the device, the processor is configured to determine at least one of a first plurality of points among the plurality of points, wherein the first plurality of points are included in a first plurality of grids, and obtain a first grid map based on the first grid including at least one of the first plurality of points, wherein the first grid map corresponds to a back surface of an external object.

[0011] In the device, the processor is configured to determine, in a first grid map including multiple rows and multiple columns, a first maximum value of the maximum value of a first plurality of points included in each column of the multiple columns, and determine a first minimum value of the minimum value of the first plurality of points, and obtain a first grid map indicating the rear contour of an external object based on the first maximum value and the first minimum value.

[0012] In the device, wherein the processor is configured to determine a first upper grid corresponding to a first maximum value, a first lower grid corresponding to a first minimum value, and a first intermediate grid included between the first upper grid and the first lower grid, determine a first reference grid included in a first reference grid map in a reference grid map, wherein the first reference grid map represents a rear surface of a vehicle, determine whether to perform a first upper match between the first upper grid and the first reference grid, determine whether to perform a first lower match between the first lower grid and the first reference grid, determine whether to perform a first intermediate match between the first intermediate grid and the first reference grid, and determine the type of the external object based on the determination of whether to perform the first upper match, whether to perform the first lower match, and whether to perform the first intermediate match.

[0013] In the device, the processor is configured to determine at least one of a second plurality of points among the plurality of points, wherein the second plurality of points are included in a second plurality of grids, and obtain a second grid map based on the second grid including at least one of the second plurality of points, wherein the second grid map corresponds to a side surface of the external object.

[0014] In the device, the processor is configured to determine, in a second grid map including multiple rows and multiple columns, a second maximum value of the maximum value of a second plurality of points included in each column of the multiple columns, determine a second minimum value of the minimum value of the second plurality of points, and obtain a second grid map based on the second maximum value and the second minimum value, wherein the second grid map represents a side profile of an external object.

[0015] In the device, the processor is configured to determine a second upper grid corresponding to a second maximum value, a second lower grid corresponding to a second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; determine a second reference grid included in a second reference grid map in a reference grid map, wherein the second grid map represents a side surface of a vehicle; determine whether to perform a second upper match between the second upper grid and the second reference grid, determine whether to perform a second lower match between the second lower grid and the second reference grid, determine whether to perform a second intermediate match between the second intermediate grid and the second reference grid; and determine the type of the external object based on the determination of whether to perform the second upper match, whether to perform the second lower match, and whether to perform the second intermediate match.

[0016] In the device, the processor is configured to assign a first score to the first grid map, wherein the first score is determined based on the first grid map and a first reference grid map included in the reference grid map; assign a second score to the second grid map, wherein the second score is determined based on the second grid map and a second reference grid map included in the reference grid map; and determine the type of the external object based on at least one of the first score or the second score.

[0017] In an apparatus, wherein a processor is configured to assign a first score to a first grid map based on detecting an external object in the same lane as a main vehicle including the apparatus; determine a type of the external object based on the first score being greater than a first reference score; assign the first score to the first grid map based on detecting an external object in a lane different from that of the main vehicle including the apparatus; assign a second score to a second grid map based on detecting an external object in a lane different from that of the main vehicle including the apparatus, and determine the type of the external object or temporarily store information indicating the type of the external object based on a total score obtained by adding the first score and the second score being greater than a second reference score greater than the first reference score.

[0018] According to the present disclosure, a method executed by a processor may include projecting multiple points indicating an external object onto a first surface to obtain a first grid map including a first plurality of grids, wherein the external object is obtained based on sensing information obtained by a sensor, and the first surface is formed by a second axis and a third axis among a first axis, a second axis and a third axis, wherein the first axis is parallel to a driving direction of the vehicle and perpendicular to the second axis, and the third axis is perpendicular to the first axis and the second axis; projecting multiple points indicating an external object onto a second surface to obtain a second grid map including a second plurality of grids, wherein the second surface is formed by the first axis and the third axis; determining the type of the external object based on at least one of the reference grid maps stored in a memory and at least one of the first grid map and the second grid map, wherein the reference grid map represents the appearance of the vehicle; and outputting a signal representing the type of the external object.

[0019] The method may further include obtaining at least one of the first grid map or the second grid map based on the external object being determined to be in a moving state.

[0020] The method may further include determining at least one of a first plurality of points among the plurality of points, wherein the first plurality of points are included in a first plurality of grids; and obtaining a first grid map based on the first grid including at least one of the first plurality of points, wherein the first grid map corresponds to a back surface of the external object.

[0021] The method may also include, in a first grid map including multiple rows and multiple columns, obtaining a first maximum value including the maximum value of a first plurality of points included in each column of the multiple columns; obtaining a first minimum value including the minimum value of the first plurality of points; and obtaining a first grid map indicating the rear contour of the external object based on the first maximum value and the first minimum value.

[0022] The method may further include determining a first upper grid corresponding to a first maximum value, a first lower grid corresponding to a first minimum value, and a first intermediate grid between the first upper grid and the first lower grid; determining a first reference grid included in a first reference grid map in a reference grid map, wherein the first reference grid map represents a rear surface of a vehicle; determining whether to perform a first upper match between the first upper grid and the first reference grid; determining whether to perform a first lower match between the first lower grid and the first reference grid; determining whether to perform a first intermediate match between the first intermediate grid and the first reference grid; and determining the type of the external object based on the determination of whether to perform the first upper match, whether to perform the first lower match, and whether to perform the first intermediate match.

[0023] The method may further include determining at least one of a second plurality of points among the plurality of points, wherein the second plurality of points are included in a second plurality of grids; and obtaining a second grid map based on a second grid including at least one of the second plurality of points, wherein the second grid map corresponds to a side surface of the external object.

[0024] The method may further include determining, in a second grid map comprising multiple rows and multiple columns, a second maximum value of the maximum value of the second multiple points included in each column of the multiple columns; determining a second minimum value of the minimum value of the second multiple points; and obtaining a second grid map based on the second maximum value and the second minimum value, wherein the second grid map represents a side profile of the external object.

[0025] The method may further include: determining a second upper grid corresponding to the second maximum value, a second lower grid corresponding to the second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; determining a second reference grid included in a second reference grid map in a reference grid map, wherein the second reference grid map represents a side surface of the vehicle; determining whether to perform a second upper match between the second upper grid and the second reference grid; determining whether to perform a second lower match between the second lower grid and the second reference grid; determining whether to perform a second intermediate match between the second intermediate grid and the second reference grid; and determining the type of the external object based on the determination of whether to perform the second upper match, whether to perform the second lower match, and whether to perform the second intermediate match.

[0026] The method may further include assigning a first score to the first grid map, wherein the first score is determined based on the first grid map and a first reference grid map included in the reference grid map; assigning a second score to the second grid map, wherein the second score is determined based on the second grid map and a second reference grid map included in the reference grid map, and determining the type of the external object based on at least one of the first score or the second score.

[0027] The method may also include: assigning a first score to a first grid map based on an external object being detected in the same lane as a main vehicle including the processor; and determining the type of the external object based on the first score being greater than a first reference score, or assigning a first score to the first grid map based on an external object being detected in a different lane from the main vehicle including the processor, assigning a second score to a second grid map, and determining the type of the external object or temporarily storing the type of the external object based on a total score obtained by adding the first score to the second score being greater than a second reference score that is greater than the first reference score. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description in conjunction with the accompanying drawings:

[0029] Figure 1 An embodiment of a block diagram of an object recognition device according to an embodiment of the present disclosure is shown;

[0030] Figure 2 An embodiment associated with a reference grid map stored in a memory in an embodiment of the present disclosure is shown;

[0031] Figure 3 An embodiment associated with a reference grid map stored in a memory in an embodiment of the present disclosure is shown;

[0032] Figure 4 An embodiment of a grid map corresponding to an external object in an embodiment of the present disclosure is shown;

[0033] Figure 5 An embodiment of comparing a grid map corresponding to an external object with a reference grid map in an embodiment of the present disclosure is shown;

[0034] Figure 6 An embodiment of a flowchart associated with an object recognition method according to another embodiment of the present disclosure is shown;

[0035] Figure 7 An embodiment of a flowchart associated with an object recognition method according to an embodiment of the present disclosure is shown;

[0036] Figure 8 An embodiment showing a result of an object recognition device according to an embodiment of the present disclosure recognizing a type of an external object; and

[0037] Fig. 9 An embodiment of a computing system including an object recognition device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0038] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the exemplary drawings. In the drawings, the same reference numerals are used throughout to refer to the same or equivalent components. In addition, detailed descriptions of well-known features or functions will be excluded to avoid unnecessarily obscuring the subject matter of the present disclosure.

[0039] When describing the components of the embodiments of the present disclosure, the terms first, second, A, B, (a), (b), etc. may be used herein. These terms are only used to distinguish one component from another component, without limiting the corresponding components, and have nothing to do with the order or priority of the corresponding components. In addition, unless otherwise defined, all terms including technical terms and scientific terms used herein will be interpreted as the customary terms of the field to which the present disclosure belongs. It should be understood that the terms used herein should be interpreted as having a meaning consistent with their meanings in the context of the present disclosure and the related art, and shall not be interpreted in an idealized or overly formal sense, unless clearly defined in this article.

[0040] In the following, reference will be made to Figures 1 to 9 Embodiments of the present disclosure are described in detail.

[0041] Figure 1 An embodiment of a block diagram of an object recognition device according to an embodiment of the present disclosure is shown.

[0042] Reference Figure 1 The object recognition device 100 according to an embodiment of the present disclosure may be implemented inside or outside a vehicle, and some components included in the object recognition device 100 may be implemented inside or outside a vehicle. In this case, the object recognition device 100 may be configured integrally with a control unit in the vehicle, or may be implemented as a separate device connected to the control unit of the vehicle through a separate connection device. For example, the object recognition device 100 may further include Figure 1 Components not shown.

[0043] Reference Figure 1 , the object recognition device 100 according to the embodiment may include a processor 110, a light detection and ranging device (LiDAR) 120, and a memory 130. The processor 110, the LiDAR 120, and the memory 130 may be electronically or operatively coupled to each other through electronic components including a communication bus.

[0044] Hereinafter, the hardware blocks are operably coupled to each other, and the coupling may include establishing a direct connection or an indirect connection between the hardware blocks in a wired or wireless manner, such that a second hardware is controlled by a first hardware in the hardware blocks.

[0045] Different blocks are shown, but the embodiments are not limited thereto. Figure 1 Some of the hardware blocks of the object recognition device 100 may be included in a single integrated circuit including a system on a chip (SoC). The types of hardware blocks included in the object recognition device 100 or the number of hardware blocks are limited to Figure 1 For example, the object recognition device 100 may include Figure 1 Some of the hardware blocks shown in .

[0046] The object recognition device 100 according to the embodiment may include hardware for processing data based on one or more instructions. The hardware for processing data may include a processor 110. For example, the hardware for processing data may include an arithmetic and logic unit (ALU), a floating point unit (FPU), a field programmable gate array (FPGA), a central processing unit (CPU) and / or an application processor (AP). The processor 110 may have a structure of a single-core processor, or may have a structure of a multi-core processor including a dual-core, quad-core, hexa-core or octa-core.

[0047] The LiDAR 120 included in the object recognition device 100 according to the embodiment may obtain a data set for recognizing surrounding objects of the object recognition device 100. For example, based on a pulse laser signal radiated from the LiDAR 120 being reflected back from surrounding objects, the LiDAR 120 may recognize at least one of the position of the surrounding objects, the movement direction of the surrounding objects, or the speed of the surrounding objects, or any combination thereof.

[0048] The memory 130 included in the object recognition device 100 according to the embodiment may include a hardware component for storing data and / or instructions input and / or output from the processor 110 of the object recognition device 100 .

[0049] For example, the memory 130 may include a volatile memory including a random access memory (RAM) or a nonvolatile memory including a read only memory (ROM).

[0050] For example, the volatile memory may include at least one of a dynamic RAM (DRAM), a static RAM (SRAM), a cache RAM, or a pseudo SRAM (PSRAM), or any combination thereof.

[0051] For example, the non-volatile memory may include at least one of a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), flash memory, a hard disk, a compact disk, a solid-state drive (SSD), or an embedded multimedia card (eMMC), or any combination thereof.

[0052] In an embodiment, the processor 110 of the object recognition device 100 may obtain a plurality of points indicating an external object obtained by the LiDAR 120. The processor 110 may project a plurality of points indicating an external object obtained by the LiDAR 120 onto a surface formed by two of the x-axis, the y-axis, and the z-axis. For example, the x-axis may include a first axis. For example, the y-axis may include a second axis. For example, the z-axis may include a third axis. Hereinafter, for ease of description, the first axis is referred to as the x-axis, the second axis is referred to as the y-axis, and the third axis is referred to as the z-axis. However, the first axis, the second axis, and / or the third axis are not limited to the above embodiments.

[0053] For example, the processor 110 may project a plurality of points indicating an external object obtained by the LiDAR 120 onto a first surface formed by the y-axis and the z-axis of the x-axis, the y-axis, and the z-axis to obtain a first grid map including a first plurality of grids. For example, the first grid map may include a grid map corresponding to a rear surface of the external object. For example, the grid map may include a plurality of grids formed by a plurality of rows and a plurality of columns. For example, the plurality of grids may include a plurality of grid cells.

[0054] For example, the processor 110 may project a plurality of points indicating an external object obtained by the LiDAR 120 onto a second surface formed by the x-axis, the y-axis, and the z-axis to obtain a second grid map including a second plurality of grids. For example, the second grid map may include a grid map corresponding to a side surface of the external object.

[0055] In an embodiment, the processor 110 may identify a reference grid map representing the appearance of the vehicle stored in the memory 130. The processor 110 may identify whether at least one of the reference grid maps representing the appearance of the vehicle stored in the memory 130 matches at least one of the first grid map and the second grid map, or any combination thereof.

[0056] In an embodiment, the processor 110 may identify the type of the external object based on matching at least one of the reference grid maps representing the appearance of the vehicle stored in the memory 130 with at least one of the first grid map and the second grid map or any combination thereof.

[0057] For example, the type of the external object can be associated with the type of the vehicle. For example, the type of the external object can include at least one of a sedan, a large truck, a 1-ton truck, a sport utility vehicle (SUV), a van, a 1-ton box truck, a trailer, or a large truck, or any combination thereof.

[0058] In an embodiment, the processor 110 may obtain a grid map for identifying the type of the external object identified by the LiDAR 120 based on the external object being identified as a moving state among a stationary state and a moving state. Based on identifying the external object as a moving state using the LiDAR 120, the processor 110 may obtain a grid map for identifying the type of the external object being tracked.

[0059] In an embodiment, the processor 110 may obtain at least one of the first grid map and the second grid map or any combination thereof based on identifying an external object within a specified range. For example, the specified range may include a distance of approximately 20 m from the front of the vehicle including the object recognition device 100. For example, the specified range may include a distance of approximately 15 m from one side of the vehicle including the object recognition device 100. For example, the processor 110 of the object recognition device 100 may obtain at least one of the first grid map or the second grid map for identifying the type of the external object or any combination thereof based on identifying that the external object is within the specified range.

[0060] Hereinafter, a description will be given of an embodiment of recognizing the type of an external object based on projecting a plurality of points onto a first surface formed by a y-axis and a z-axis to obtain a first mesh map including a first plurality of meshes.

[0061] In an embodiment, the processor 110 may identify at least one (e.g., all or some) of the first plurality of points included in the first plurality of grids among the plurality of points. For example, the first plurality of points may be obtained based on projecting the plurality of points obtained by the LiDAR 120 onto the first surface.

[0062] In an embodiment, the processor 110 may identify a first grid including all or some of the first plurality of points in the first plurality of grids. For example, the first grid may include a grid in which the first plurality of points are identified. The processor 110 may obtain a first grid map corresponding to the rear surface of the external object based on the first grid including all or some of the first plurality of points in the first plurality of grids.

[0063] In an embodiment, the processor 110 may identify a maximum value of a first plurality of points included in each of a plurality of columns in a plurality of rows and a plurality of columns included in the first grid map. For example, the processor 110 may identify a first maximum value including a maximum value of a first plurality of points, the first plurality of points included in each of the plurality of columns.

[0064] In an embodiment, the processor 110 may identify a minimum value of a first plurality of points included in each of the plurality of columns. The processor 110 may identify a first minimum value including a minimum value of a first plurality of points included in each of the plurality of columns.

[0065] In an embodiment, the processor 110 may obtain a first grid map indicating a rear contour of the external object based on the first maximum value and the first minimum value.

[0066] In an embodiment, the processor 110 may identify a first upper grid corresponding to a first maximum value. The processor 110 may identify a first lower grid corresponding to a first minimum value. The processor 110 may identify a first intermediate grid included between the first upper grid and the first lower grid. As described above, the processor 110 may identify a first upper grid, a first lower grid, and a first intermediate grid.

[0067] In an embodiment, the processor 110 may identify a first reference grid included in the first reference grid map in the reference grid map included in the memory 130, the first reference grid representing the rear surface of the vehicle. Figure 2 and Figure 3 Description A description associated with the first reference grid diagram.

[0068] In an embodiment, the processor 110 may identify a first reference grid included in the first reference grid map. For example, the first reference grid may include a grid included in the first reference grid map for representing a rear surface of the vehicle.

[0069] In an embodiment, the processor 110 may determine whether to perform a first upper match between the first upper grid and the first reference grid. For example, the processor 110 may determine whether the first upper grid and the upper grid included in the first reference grid map match each other. Whether to perform the first upper match may be associated with an operation of determining whether the first upper grid and the upper grid included in the first reference grid map match each other.

[0070] In an embodiment, the processor 110 may determine whether to perform a first lower match between the first lower grid and the first reference grid. For example, the processor 110 may determine whether the first lower grid and the lower grid included in the first reference grid map match each other. Whether to perform the first lower match may be associated with an operation of determining whether the first lower grid and the lower grid included in the first reference grid map match each other.

[0071] In an embodiment, the processor 110 may determine whether to perform a first intermediate match between the first intermediate grid and the first reference grid. For example, the processor 110 may determine whether the first intermediate grid and the intermediate grid included in the first reference grid map match each other. Whether to perform the first intermediate match may be associated with an operation of determining whether the first intermediate grid and the intermediate grid included in the first reference grid map match each other.

[0072] In an embodiment, the processor 110 may identify the type of the external object based on the number of matches and the number of mismatches calculated by performing the first upper match, performing the first lower match, and performing the first intermediate match. For example, the number of matches calculated by performing the first upper match may include the number of matches between the first upper grid and the upper grid included in the first reference grid map. For example, the number of matches calculated by performing the first lower match may include the number of matches between the first lower grid and the lower grid included in the first reference grid map. For example, the number of matches calculated by performing the first intermediate match may include the number of matches between the first intermediate grid and the intermediate grid included in the first reference grid map.

[0073] For example, the number of mismatches calculated by performing the first upper matching may include the number of mismatches between the first upper grid and the upper grid included in the first reference grid map. For example, the number of mismatches calculated by performing the first lower matching may include the number of mismatches between the first lower grid and the lower grid included in the first reference grid map. For example, the number of mismatches calculated by performing the first intermediate matching may include the number of mismatches between the first intermediate grid and the intermediate grid included in the first reference grid map.

[0074] As described above, the processor 110 of the object recognition device 100 according to an embodiment may recognize the type of the external object based on the number of matches and the number of mismatches calculated by performing the first upper matching, performing the first lower matching, and performing the first middle matching.

[0075] Hereinafter, a description will be given of an embodiment of recognizing the type of an external object based on projecting a plurality of points onto a second surface formed by an x-axis and a z-axis to obtain a second mesh map including a second plurality of meshes.

[0076] In an embodiment, the processor 110 may identify all or some of the second plurality of points in the plurality of points included in the second plurality of grids. For example, the second plurality of points may be obtained based on the plurality of points obtained by the LiDAR 120 being projected onto the second surface.

[0077] In an embodiment, the processor 110 may identify a second grid including all or some of the second plurality of points in the second plurality of grids. For example, the second grid may include a grid in which the second plurality of points are identified. The processor 110 may obtain a second grid map corresponding to the side surface of the external object based on the second grid including all or some of the second plurality of points in the second plurality of grids.

[0078] In an embodiment, the processor 110 may identify a maximum value of a second plurality of points included in each of the plurality of columns, among a plurality of rows and a plurality of columns included in the second grid map. For example, the processor 110 may identify a second maximum value including a maximum value of a second plurality of points, the second plurality of points included in each of the plurality of columns.

[0079] In an embodiment, the processor 110 may identify a minimum value of a second plurality of points included in each of the plurality of columns. The processor 110 may identify a second minimum value including a minimum value of a second plurality of points included in each of the plurality of columns.

[0080] In an embodiment, the processor 110 may obtain a second grid map indicating a side profile of the external object based on the second maximum value and the second minimum value.

[0081] In an embodiment, the processor 110 may identify a second upper grid corresponding to the second maximum value. The processor 110 may identify a second lower grid corresponding to the second minimum value. The processor 110 may identify a second intermediate grid included between the second upper grid and the second lower grid. As described above, the processor 110 may identify the second upper grid, the second lower grid, and the second intermediate grid.

[0082] In an embodiment, the processor 110 may identify the second reference grid included in the second reference grid map representing the rear surface of the vehicle in the reference grid map included in the memory 130. Figure 2 Description A description associated with the second reference grid diagram.

[0083] In an embodiment, the processor 110 may identify a second reference grid included in the second reference grid map. For example, the second reference grid may include a grid included in the second reference grid map for representing a side surface of the vehicle.

[0084] In an embodiment, the processor 110 may determine whether to perform a second upper match between the second upper grid and the second reference grid. For example, the processor 110 may determine whether the second upper grid and the upper grid included in the second reference grid map match each other. Whether to perform the second upper match may be associated with an operation of determining whether the second upper grid and the upper grid included in the second reference grid map match each other.

[0085] In an embodiment, the processor 110 may determine whether to perform a second lower match between the second lower grid and the second reference grid. For example, the processor 110 may determine whether the second lower grid and the lower grid included in the second reference grid map match each other. Whether to perform the second lower match may be associated with an operation of determining whether the second lower grid and the lower grid included in the second reference grid map match each other.

[0086] In an embodiment, the processor 110 may determine whether to perform a second intermediate matching between the second intermediate grid and the second reference grid. For example, the processor 110 may identify whether the second intermediate grid and the intermediate grid included in the second reference grid map match each other. Whether to perform the second intermediate matching may be associated with an operation of determining whether the second intermediate grid and the intermediate grid included in the second reference grid map match each other.

[0087] In an embodiment, the processor 110 may identify the type of the external object based on the number of matches and the number of mismatches calculated by performing the second upper match, performing the second lower match, and performing the second intermediate match. For example, the number of matches calculated by performing the second upper match may include the number of matches between the second upper grid and the upper grid included in the second reference grid map. For example, the number of matches calculated by performing the second lower match may include the number of matches between the second lower grid and the lower grid included in the second reference grid map. For example, the number of matches calculated by performing the second intermediate match may include the number of matches between the second intermediate grid and the intermediate grid included in the second reference grid map.

[0088] For example, the number of mismatches calculated by performing the second upper matching may include the number of mismatches between the second upper grid and the upper grid included in the second reference grid map. For example, the number of mismatches calculated by performing the second lower matching may include the number of mismatches between the second lower grid and the lower grid included in the second reference grid map. For example, the number of mismatches calculated by performing the second intermediate matching may include the number of mismatches between the second intermediate grid and the intermediate grid included in the second reference grid map.

[0089] As described above, the processor 110 of the object recognition device 100 according to an embodiment may recognize the type of the external object based on the number of matches and the number of mismatches calculated by performing the second upper matching, performing the second lower matching, and performing the second middle matching.

[0090] In an embodiment, the processor 110 may determine the first matching number and the first mismatching number by performing a first upper matching, a first lower matching, and a first intermediate matching. The processor 110 may determine the second matching number and the second mismatching number by performing a second upper matching, a second lower matching, and a second intermediate matching.

[0091] For example, the processor 110 may determine the first score based on the first reference grid included in the first reference grid map, the first matching number, and the first mismatching number. The processor 110 may identify the type of the external object based on the first score.

[0092] For example, the processor 110 may determine the second score based on the second reference grid included in the second reference grid map, the second matching number, and the second mismatching number. The processor 110 may identify the type of the external object based on the second score.

[0093] As described above, the object recognition device 100 according to the embodiment may recognize the maximum value and the minimum value in each of a plurality of columns of a grid graph and may recognize the type of an external object, thereby increasing calculation speed and reducing the amount of calculation of the processor 110 .

[0094] Figure 2 An embodiment associated with a reference grid map stored in a memory in an embodiment of the present disclosure is shown.

[0095] refer to Figure 2 The described processing can be done by Figure 1 Alternatively, referring to Figure 2 The described processing can be performed by including Figure 1 The processor 110 in the object recognition device 100 executes.

[0096] refer to Figure 2 , the first reference grid map 201 may include a template representing the rear surface of the vehicle based on a first surface formed by the y-axis and the z-axis among the x-axis (e.g., parallel to the driving direction of the vehicle), the y-axis (e.g., perpendicular to the x-axis), and the z-axis (e.g., perpendicular to both the x-axis and the y-axis). The second reference grid map 203 may include a template representing the side surface of the vehicle based on a second surface formed by the x-axis and the z-axis among the x-axis, the y-axis, and the z-axis.

[0097] In an embodiment, a memory included in the object recognition device may include a first reference grid map 201 and a second reference grid map 203 representing the appearance of a vehicle.

[0098] For example, the first reference grid diagram 201 may include a plurality of grids. For example, the first reference grid diagram 201 may include a plurality of grids divided by a plurality of columns and a plurality of rows. For example, each first grid diagram 201 may include about 30 rows and about 50 columns.

[0099] For example, the second reference grid diagram 203 may include a plurality of grids. For example, the second reference grid diagram 203 may include a plurality of grids divided by a plurality of rows and a plurality of columns. For example, each second grid diagram 203 may include about 30 rows and about 100 columns.

[0100] The template described below is an embodiment of the first reference grid map 201 and the second reference grid map 203 and is not limited by name.

[0101] For example, the first templates 211 and 221 may include a template representing the appearance of a sedan. For example, the first template 211 may include a template representing the rear surface of a sedan. For example, the first template 221 may include a template representing the side surface of a sedan.

[0102] For example, the second templates 212 and 222 may include a template representing the appearance of a truck having a first size. For example, the truck having the first size may include a truck having a size larger than a truck having a second size and a truck having a third size, which will be described below. For example, the second template 212 may include a template representing the rear surface of the truck having the first size. For example, the second template 222 may include a template representing the side surface of the truck having the first size.

[0103] For example, the third templates 213 and 223 may include a template representing the appearance of a truck having a second size. For example, the truck having the second size may include a truck having a size smaller than the truck having the first size and the truck having the third size. For example, the truck having the second size may include a 1-ton truck. For example, the third template 213 may include a template representing the rear surface of the truck having the second size. For example, the third template 223 may include a template representing the side surface of the truck having the second size.

[0104] For example, the fourth templates 214 and 224 may include a template representing the exterior of a van or a sport utility vehicle (SUV). For example, the fourth template 214 may include a template representing the rear surface of a van or an SUV. For example, the fourth template 224 may include a template representing the side surface of a van or an SUV.

[0105] For example, the fifth templates 215 and 225 may include a template representing the appearance of a truck in which a box is coupled to a truck having a second size. For example, the fifth template 215 may include a template representing the rear surface of a truck in which a box is coupled to a truck having a second size. For example, the fifth template 225 may include a template representing the side surface of a truck in which a box is coupled to a truck having a second size.

[0106] For example, the sixth templates 216 and 226 may include a template representing the appearance of a truck connected to a trailer. For example, the sixth template 216 may include a template representing the rear surface of a truck connected to a trailer. For example, the sixth template 226 may include a template representing the side surface of a truck connected to a trailer.

[0107] For example, the seventh templates 217 and 227 may include a template representing the appearance of a truck having a third size. For example, the truck having the third size may include a truck that is smaller than the truck having the first size and larger than the truck having the second size. For example, the truck having the third size may include a 2.5 ton truck. For example, the seventh template 217 may include a template representing the rear surface of the truck having the third size. For example, the seventh template 227 may include a template representing the side surface of the truck having the third size.

[0108] As described above, the template may include the grid included in Table 1 below.

[0109] [Table 1]

[0110]

[0111] Referring to Table 1 above, the type may include the type of the external object. The types described with reference to Table 1 above may correspond to the first template 211 to the seventh template 227 in sequence. For example, the first templates 211 and 221 may include a template representing a sedan. For example, the second templates 212 and 222 may include a template representing a large truck. For example, the third templates 213 and 223 may include a template representing a 1-ton truck. For example, the fourth templates 214 and 224 may include a template representing a Bongo. For example, the fifth templates 215 and 225 may include a template representing a 1-ton box truck. For example, the sixth templates 216 and 226 may include a template representing a trailer. For example, the seventh templates 217 and 227 may include a template representing a large truck.

[0112] In the above Table 1, the representation grid may include at least one of a first reference grid included in the first reference grid map or a second reference grid included in the second reference grid map or any combination thereof. In the above Table 1, the background grid may include a grid that does not indicate a type of vehicle.

[0113] In an embodiment, the processor of the object recognition device may include Figure 2 The processor may compare at least one of the expression grid or the background grid in the corresponding reference grid map (e.g., the first reference grid map 201 and the second reference grid map 203) or any combination thereof with the grid map corresponding to the external object, and may identify the type of the external object. For example, the processor may compare the first reference grid map 201 with the first grid map indicating the rear surface of the external object, and may identify the type of the external object. For example, the processor may compare the second reference grid map 203 with the second grid map indicating the side surface of the external object, and may identify the type of the external object.

[0114] Figure 3An embodiment associated with a reference grid map stored in a memory in an embodiment of the present disclosure is shown.

[0115] Reference Figure 3 The described processing can be done by Figure 1 Alternatively, referring to Figure 3 The described processing can be performed by including Figure 1 The processor 110 in the object recognition device 100 executes.

[0116] refer to Figure 3 , Figure 3 The first reference grid map 301 may include a plurality of grids representing Figure 2 An embodiment of the first template 211. Figure 3 The second reference grid map 303 may include a plurality of grids representing Figure 2 An embodiment of the first template 221.

[0117] refer to Figure 3 , the first reference grid map 301 and the second reference grid map 303 may include reference grids representing the vehicle. For example, the first reference grid map 301 may include a first reference grid representing the rear surface of the vehicle. For example, the second reference grid map 303 may include a second reference grid representing the side surface of the vehicle.

[0118] For example, the first reference grid map 301 may include a first expression grid. For example, the first reference grid map 301 may include a first background grid.

[0119] For example, the second reference grid map 303 may include a second expression grid. For example, the second reference grid map 303 may include a second background grid.

[0120] In an embodiment, the processor of the object recognition device may compare the first grid map indicating the rear surface of the external object with the first reference grid map 301. For example, the processor may identify a first maximum value, the first maximum value including a maximum value identified from each of a plurality of columns included in the first grid map indicating the rear surface of the external object. For example, the processor may identify a first minimum value, the first minimum value including a minimum value identified from each of a plurality of columns included in the first grid map indicating the rear surface of the external object. For example, the processor may identify a first upper grid corresponding to the first maximum value. For example, the processor may identify a first lower grid corresponding to the first minimum value. The processor may identify a first intermediate grid included between the first upper grid and the first lower grid.

[0121] In an embodiment, the processor may identify whether the first upper grid, the first lower grid, and the first middle grid match the first reference grid included in the first reference grid map 301. The processor may compare the first upper grid, the first lower grid, and the first middle grid with the first expression grid and the first background grid. The processor may identify the type of the external object based on the comparison of the first upper grid, the first lower grid, and the first middle grid with the first expression grid and the first background grid.

[0122] In an embodiment, the processor of the object recognition device may compare the second grid map indicating the side surface of the external object with the second reference grid map 303. For example, the processor may identify a second maximum value, the second maximum value including a maximum value identified from each of a plurality of columns included in the second grid map indicating the side surface of the external object. For example, the processor may identify a second minimum value, the second minimum value including a minimum value identified from each of a plurality of columns included in the second grid map indicating the rear surface of the external object. For example, the processor may identify a second upper grid corresponding to the second maximum value. For example, the processor may identify a second lower grid corresponding to the second minimum value. The processor may identify a second intermediate grid included between the first upper grid and the first lower grid.

[0123] In an embodiment, the processor may identify whether the second upper grid, the second lower grid, and the second middle grid match the second reference grid included in the second reference grid map 303. The processor may compare the second upper grid, the second lower grid, and the second middle grid with the second expression grid and the second background grid. The processor may identify the type of the external object based on the comparison of the second upper grid, the second lower grid, and the second middle grid with the second expression grid and the second background grid.

[0124] Figure 4 An embodiment of a grid map corresponding to an external object in an embodiment of the present disclosure is shown.

[0125] refer to Figure 4 The described processing can be done by Figure 1 Alternatively, referring to Figure 4 The described processing can be performed by including Figure 1 The processor 110 in the object recognition device 100 executes.

[0126] Reference Figure 4 , the processor of the object recognition device according to the embodiment can detect the external object using LiDAR. For example, the processor can identify a plurality of points indicating the external object obtained by LiDAR. For example, the external object can be represented by a plurality of points obtained by LiDAR as the first embodiment 410.

[0127] In an embodiment, the processor may project the plurality of points onto a surface formed by two of an x-axis (e.g., a driving direction of the vehicle), a y-axis (e.g., perpendicular to the x-axis), and a z-axis (e.g., perpendicular to both the x-axis and the y-axis). The processor may project the plurality of points onto a first surface 411 formed by the y-axis and the z-axis among the x-axis, the y-axis, and the z-axis. The processor may identify the first plurality of points based on projecting the plurality of points onto the first surface 411.

[0128] In an embodiment, the processor may identify a plurality of rows and a plurality of columns including a plurality of grids on a first surface 411, such as a second embodiment 420. The processor may identify a maximum value and a minimum value from each of the plurality of columns. The processor may identify a first maximum value 421, the first maximum value including the maximum value identified from each of the plurality of columns. The processor may identify a first minimum value 422, the first minimum value including the minimum value identified from each of the plurality of columns.

[0129] Referring to the third embodiment 430, the processor may identify first upper grids 451 respectively corresponding to the first maximum values ​​421. The processor may identify first lower grids 453 respectively corresponding to the first minimum values ​​422. The processor may identify a first intermediate grid 455 included between the first upper grid 451 and the first lower grid 453. The processor may obtain a first grid map 431 including the first upper grid 451, the first lower grid 453, and the first intermediate grid 455.

[0130] In an embodiment, the processor may project the plurality of points onto a surface formed by two of the x-axis, the y-axis, and the z-axis. The processor may project the plurality of points onto a second surface 413 formed by the x-axis and the z-axis of the x-axis, the y-axis, and the z-axis. The processor may identify the second plurality of points based on projecting the plurality of points onto the second surface 413.

[0131] In an embodiment, similar to the second embodiment 420, the processor may identify a plurality of rows and a plurality of columns including a plurality of grids on the second surface 413. The processor may identify a maximum value and a minimum value from each of the plurality of columns. The processor may identify a second maximum value 423, the second maximum value 423 including the maximum value identified from each of the plurality of columns. The processor may identify a second minimum value 424, the second minimum value including the minimum value identified from each of the plurality of columns.

[0132] Referring to the third embodiment 430, the processor may identify second upper grids 451 respectively corresponding to the second maximum values ​​423. The processor may identify second lower grids 453 respectively corresponding to the second minimum values ​​424. The processor may identify a second intermediate grid 455 included between the second upper grid 451 and the second lower grid 453. The processor may obtain a second grid map 433 including the second upper grid 451, the second lower grid 453, and the second intermediate grid 455.

[0133] In the following, embodiments for identifying an upper grid and a lower grid will be described with reference to Table 2 below, Equation 1 below, and Table 3 below.

[0134] [Table 2]

[0135]

[0136] The above Table 2 may include instructions (or program codes) for identifying maximum and minimum values from each of multiple columns using multiple points indicating an external object at the current time point. The above Table 2 may include instructions (or program codes) for calculating an occupancy count of multiple points indicating an external object, which accumulates from a past time point to the current time point.

[0137] For example, if(YZFace[col].curMaxRow<row_input)YZFace[col].curMaxRow=row_input in the above Table 2 may include instructions for identifying a maximum value on a first surface formed by the y-axis and the z-axis at the current time point.

[0138] For example, if(YZFace[col].curMinRow<row_input YZFace[col].curMaxRow<row_input in the above Table 2 may include instructions for identifying a minimum value on a first surface formed by the y-axis and the z-axis at the current time point.

[0139] For example, if(YZFace[col].curCnt<PIXEL_SIZE)YZFace[col].curCnt++ in the above Table 2 may include instructions for identifying whether the size of a first grid map obtained at the current time point matches the size of a first reference grid map.

[0140] For example, if(XZFace[col].curMaxRow<row_input)XZFace[col].curMaxRow=row_input in the above Table 2 may include instructions for identifying a maximum value on a second surface formed by the x-axis and the z-axis at the current time point.

[0141] For example, if(XZFace[col].curMinRow<row_input XZFace[col].curMaxRow<row_input in the above Table 2 may include instructions for identifying a minimum value on a second surface formed by the x-axis and the z-axis at the current time point.

[0142] For example, the instruction if(XZFace[col].curCnt<PIXEL_SIZE)XZFace[col].curCnt++ in Table 2 above may include an instruction for identifying whether the size of the second grid map obtained at the current time point matches the size of the second reference grid map.

[0143] [Equation 1]

[0144]

[0145] [Table 3]

[0146]

[0147] Referring to Equation 1 above, the processor may use YZFace[col].curCnt obtained from Table 1 above to obtain Average YZFace . For example, Average YZFace may include an average value of a grid for identifying a plurality of points indicating the rear surface of an external object from each of a plurality of columns included in the first grid map. NumOffCol in Equation 1 above may include the number of columns included in the grid map. PIXEL_COL_SIZE in Equation 1 above may include the number of columns included in the reference grid map. YZFace[col].curCnt in Equation 1 above may include the number of columns that include a grid indicating an external object at the current time point.

[0148] Referring to Table 3 above, based on YZFace[col].curCnt being less than Average YZFace , the processor may output YZFace[col].curMaxRow as "0" and may output YZFace[col].MaxRow as "0". Based on YZFace[col].curCnt being greater than or equal to Average YZFace , the processor may output YZFace[col].MaxRow as YZFace[col].curMaxRow and may output YZFace[col].MinRow as YZFace[col].curMinRow.

[0149] Hereinafter, embodiments of comparing a obtained grid map (e.g., at least one of the first grid map 431 or the second grid map 433 or any combination thereof) with a reference grid map will be described.

[0150] Figure 5 Embodiments of comparing a grid map corresponding to an external object with a reference grid map in embodiments of the present disclosure are shown.

[0151] refer to Figure 5 The described processing can be done by Figure 1 Alternatively, referring to Figure 5 The described processing can be performed by including Figure 1 The processor 110 in the object recognition device 100 executes.

[0152] Reference Figure 5 , the processor of the object recognition device according to the embodiment may compare the grid map 520 with the reference grid map 510. For example, the object recognition device may perform a combination 530 of the grid included in the grid map 520 and the reference grid included in the reference grid map 510. Hereinafter, the operation of assigning a score may be included in the operation of calculating a score.

[0153] For example, the processor may perform a combination 530 of the first reference grid 511 included in the reference grid map 510 and the first grid 521 included in the grid map 520. For example, the processor may determine whether the first reference grid 511 and the first grid 521 match each other. Based on the matching of the first reference grid 511 and the first grid 521, the processor may output "1" (e.g., Boolean true). For example, the "1" output based on the matching of the first reference grid 511 and the first grid 521 may be included in the matching number.

[0154] For example, the processor may perform a combination 530 of the second reference grid 513 included in the reference grid map 510 and the second grid 523 included in the grid map 520. For example, the processor may identify whether the second reference grid 513 and the second grid 523 match each other. Based on the second reference grid 513 and the second grid 523 matching each other, the processor may output "1". For example, the "1" output based on the second reference grid 513 and the second grid 523 matching each other may be included in the matching number.

[0155] For example, the processor may perform a combination 530 of the third reference grid 515 included in the reference grid map 510 and the third grid 525 included in the grid map 520. For example, the processor may identify whether the third reference grid 515 and the third grid 525 match each other. Based on the fact that the third reference grid 515 and the third grid 525 do not match each other, the processor may output "0", (e.g., Boolean value false). For example, the "0" output based on the fact that the third reference grid 515 and the third grid 525 do not match each other may be included in the mismatch number.

[0156] The processor may assign a score to the grid map 520 based on the number of matches and the number of non-matches output by comparing the reference grid map 510 with the grid map 520. The processor may identify the type of the external object based on the score assigned to the grid map 520.

[0157] For example, the processor may assign the score using Equation 2 described below.

[0158] [Equation 2]

[0159]

[0160] Referring to Equation 2 above, GFI is the abbreviation of goodness of fit index, which refers to the suitability. Figure 3 Described expression grid with reference Figure 4 At least one of the upper grid, lower grid or middle grid described above or any combination thereof matches each other to output the resultOfValue of equation 2 above. Figure 3 The background grid described and the reference Figure 4 The resultOfNull of the above equation 2 may be output based on the fact that the grids included in the reference grid map and the grids included in the grid map do not match each other. The AnsSizeYZ of the above equation 2 may include the number of expression grids included in the first reference grid map represented on the y-axis and the z-axis. The NullSizeYZ of the above equation 2 may include the number of background grids included in the first reference grid map represented on the y-axis and the z-axis.

[0161] For example, the processor may use the instruction output including the following Table 5 to output at least one of resultOfValue++, resultOfNull++, or resultOfError++, or any combination thereof.

[0162] [Table 5]

[0163]

[0164] Table 5 above may include instructions for comparing a first surface formed by the y-axis and the z-axis with a first reference grid included in a first reference grid map.

[0165] For example, for (col < PIXEL_COL_SIZE) and for (row < PIXEL_ROW_SIZE) may include instructions for sequentially comparing multiple rows and multiple columns. For example, if the first grid matches the first reference grid, where the first grid is included in the first grid diagram and corresponds to the first reference grid included in the first reference grid diagram, if (AnsYZFace[col][row] == 1 && TestYZFace[col][row] == 1) and resultOfValue++ may include instructions for outputting resultOfValue++.

[0166] For example, when the first grid included in the first grid diagram and corresponding to the first reference grid included in the first reference grid diagram does not match the first reference grid, else if (AnsYZFazce[col][row] == 0 && TestYZFace[col][row] == 0) and resultOfNull++ may include instructions for outputting resultOfNull++.

[0167] For example, if at least one or any combination of the above resultOfValue++ or resultOfNull++ is not output, else resultOfError++ may contain instructions for outputting resultOfError++.

[0168] In an embodiment, the processor may assign a score to the first grid diagram based on the instructions included in Table 5 above. The processor may use the instructions included in Table 5 above to assign a score to the first grid diagram and may identify the type of the external object corresponding to the first grid diagram based on the score assigned to the first grid diagram.

[0169] In an embodiment, the processor may assign a score to the second grid diagram using instructions in which some of the instructions included in Table 5 above are changed. For example, the processor may assign a score to the second grid diagram and may identify the type of the external object corresponding to the second grid diagram based on the score assigned to the second grid diagram.

[0170] For example, the processor may assign a first score calculated based on the first grid diagram and the first reference grid diagram included in the reference grid diagram to the first grid diagram. For example, the processor may assign a second score calculated based on the second grid diagram and the second reference grid diagram included in the reference grid diagram to the second grid diagram. The processor may identify the type of the external object based on at least one of the first score or the second score or any combination thereof.

[0171] For example, if an external object is identified in the same lane as the host vehicle including the object recognition device, the processor may assign a first score to the first grid map. The processor may identify the type of the external object based on the first score being greater than a first reference score.

[0172] For example, if an external object is recognized in a lane different from that of the host vehicle including the object recognition device, the processor may assign a first score to the first grid map and may assign a second score to the second grid map. Based on a total score obtained by adding the first score and the second score being greater than a second reference score greater than the first reference score, the processor may recognize the type of the external object or may temporarily store the type of the external object.

[0173] Figure 6 An embodiment of a flowchart associated with an object recognition method according to another embodiment of the present disclosure is shown.

[0174] In the following, it is assumed Figure 1 The object recognition device 100 performs Figure 6 In addition, Figure 6 In the description of , operations described as being performed by the device may be understood as being controlled by the processor 110 of the object recognition device 100 .

[0175] Figure 6 At least one of the operations may be performed by Figure 1 The object recognition device 100 is executed. Figure 6 The corresponding operations of can be performed sequentially, but not necessarily sequentially. For example, the order of the operations can be changed, and at least two operations can be performed in parallel.

[0176] refer to Figure 6 In S601, the object recognition method according to the embodiment may include projecting a plurality of points indicating an external object obtained by LiDAR onto a first surface formed by the y-axis and the z-axis among the x-axis, the y-axis, and the z-axis. The object recognition method may include projecting a plurality of points indicating an external object obtained by LiDAR onto the first surface to obtain a first grid map including a first plurality of grids.

[0177] Obtaining the first grid map may include referring to Figure 4 Description of the process.

[0178] In S602, the object recognition method according to the embodiment may include projecting a plurality of points indicating an external object obtained by the LiDAR onto a second surface formed by an x-axis, a y-axis, and a z-axis. The object recognition method may include projecting a plurality of points indicating an external object obtained by the LiDAR onto the second surface to obtain a second grid map including a second plurality of grids.

[0179] Obtaining the second grid map may include referring to Figure 4 Description of the process.

[0180] In S603, the object recognition method according to the embodiment may include: identifying whether at least one reference grid map among the reference grid maps representing the appearance of the vehicle stored in the memory matches at least one of the first grid map or the second grid map or any combination thereof.

[0181] For example, the object recognition method may include: identifying the type of the external object based on matching at least one reference grid map among the reference grid maps representing the appearance of the vehicle stored in the memory with at least one of the first grid map or the second grid map or any combination thereof.

[0182] Figure 7 An embodiment of a flowchart associated with an object recognition method according to an embodiment of the present disclosure is shown.

[0183] In the following, it is assumed Figure 1 The object recognition device 100 performs Figure 7 In addition, Figure 7 In the description of , operations described as being performed by the device may be understood as being controlled by the processor 110 of the object recognition device 100 .

[0184] Figure 7 At least one of the operations may be performed by Figure 1 The object recognition device 100 is executed. Figure 7 The operations of can be performed sequentially, but not necessarily sequentially. For example, the order of the operations can be changed, and at least two operations can be performed in parallel.

[0185] Reference Figure 7 In S701, the object recognition method according to the embodiment may include obtaining a first grid map based on a first surface indicating a rear surface of the external object. The object recognition method may include obtaining a second grid map based on a second surface indicating a side surface of the external object.

[0186] The object recognition method may include: calculating a score based on matching the first grid map and the second grid map with a reference grid map stored in a memory. The calculation of the score may include assigning a reference Figure 5 The score of the description.

[0187] In S703, the object recognition method according to an embodiment may include identifying a lane that the external object is traveling in. For example, the object recognition method may include determining whether the external object is identified as being in the same lane as the host vehicle.

[0188] If the external object is identified as being in the same lane as the host vehicle (Yes in S703), in S705, the object recognition method according to the embodiment may include identifying whether a score calculated based on whether the first grid map based on the first surface and the first reference grid map match each other is greater than a first reference score. For example, the first reference score may include about 70 points.

[0189] If the external object is operating in the same lane as the host vehicle, since a first plurality of points corresponding to a rear surface of the external object is obtained, the object recognition method may include recognizing a type of the external object using a first grid map based on the first surface and a first reference grid map.

[0190] When the score calculated based on whether the first grid map based on the first surface matches the first reference grid map is greater than the first reference score ("Yes" in S705), in S707, the object recognition method according to the embodiment may include determining the type of the external object. For example, when the score calculated based on whether the first grid map and the first reference grid map match each other is greater than the first reference score, the object recognition method may include identifying the type of the external object and determining the type of the external object to mark the external object with the determined type.

[0191] When the score calculated according to whether the first grid map based on the first surface and the first reference grid map match each other is less than or equal to the first reference score (No in S705 ), the object recognition method according to the embodiment may end the process of applying the present technology.

[0192] If the external object is not identified in the same lane as the main vehicle (No in S703), then in S709, the object recognition method according to the embodiment may include identifying whether the calculated score is greater than a first reference score, a second reference score greater than the first reference score, and a third reference score greater than the second reference score.

[0193] For example, the first reference score may include about 70 points. For example, the second reference score may include about 140 points. For example, the third reference score may include about 160 points.

[0194] For example, if no external object is recognized in the same lane as the host vehicle, the object recognition method may include obtaining a first score calculated based on whether a first grid map based on a first surface representing a rear surface of the external object and a first reference grid map match each other. The object recognition method may include obtaining a second score calculated based on whether a second grid map indicating a second surface of a side surface of the external object and a second reference grid map match each other.

[0195] The object recognition method may include recognizing the type of the external object based on a total score obtained by adding the obtained first score to the obtained second score. The object recognition method may include determining whether the total score is greater than a third reference score.

[0196] If the calculated score is greater than the third reference score (Yes at S709 ), the object recognition method according to the embodiment may include determining the type of the external object.

[0197] For example, if the calculated score (eg, the total score) is greater than the third reference score, the object recognition method may include recognizing a type of the external object and determining the type of the external object to mark the external object with the determined type.

[0198] If the calculated score is less than or equal to the third reference score (No in S709 ), then in S713 , the object recognition method according to the embodiment may include identifying whether the calculated score is greater than a second reference score.

[0199] If the calculated score is greater than the second reference score (Yes in S713), then in S715, the object recognition method according to the embodiment may include recording the type of the external object. For example, recording the type of the external object may include temporarily storing the type of the external object in a separate area of ​​the memory.

[0200] If the calculated score is less than or equal to the second reference score (No at S713 ), the object recognition method according to the embodiment may include ending the process of applying the present technology.

[0201] In S717, the object recognition method according to the embodiment may include selecting the type with the highest calculated score as the type of the external object at the current time point. For example, if multiple types of external objects are temporarily stored, the object recognition method may include identifying the highest score in the total score for identifying the multiple stored types. The object recognition method may include selecting the type corresponding to the highest score in the total score as the type of the external object.

[0202] Figure 8 An example of a result of the object recognition apparatus according to an embodiment of the present disclosure recognizing the type of an external object is shown.

[0203] Figure 8 The host vehicle 800 may include Figure 1 The object recognition device 100 is provided.

[0204] refer to Figure 8In the first embodiment 801, the processor of the object recognition device according to the embodiment may use a sensor (e.g., a LiDAR sensor) to recognize an external object 810. For example, the external object 810 may correspond to another vehicle (e.g., a truck including a head and a tail) or a pedestrian.

[0205] For example, when recognizing the external object 810 using LiDAR, the processor may recognize the head and tail of the external object 810 separated from each other because the processor recognizes the type of the external object 810 using a point group (e.g., point cloud). As described above, when recognizing the external object 810 whose head and tail are separated from each other, the processor may incorrectly recognize the heading direction of the external object 810.

[0206] The second embodiment 803 may include an embodiment of applying the present technology to identify the type of the external object 820. For example, the processor may determine or identify the type of the external object 820 using at least one of the first grid map or the second grid map indicating the external object 820 or any combination thereof, thereby accurately determining or identifying the external object 820. The processor may accurately identify the external object 820, thereby reducing the situation where the external object 820 is separated and identified.

[0207] Fig. 9 An embodiment of a computing system including an object recognition device according to an embodiment of the present disclosure is shown.

[0208] refer to Fig. 9 , the computing system 1000 may include at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage device 1600 , and a network interface 1700 connected to each other via a bus 1200 .

[0209] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1300 and / or the memory 1600. The memory 1300 and the storage device 1600 may include various types of volatile or non-volatile storage media. For example, the memory 1300 may include a read-only memory (ROM) 1310 and a random access memory (RAM) 1320.

[0210] Therefore, the operation of the method or algorithm described in conjunction with the embodiments disclosed in the specification can be directly implemented using a hardware module, a software module, or a combination of a hardware module and a software module, which is executed by the processor 1100. The software module may reside on a storage medium (i.e., the memory 1300 and / or the storage device 1600) such as a RAM, a flash memory, a ROM, an EPROM, an EEPROM, a register, a hard disk, a removable disk, and a CD-ROM.

[0211] An exemplary storage medium may be coupled to the processor 1100. The processor 1100 may read information from the storage medium and may write information to the storage medium. Alternatively, the storage medium may be integrated with the processor 1100. The processor and the storage medium may be present in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In another case, the processor and the storage medium may reside in a user terminal as separate components.

[0212] The present disclosure is directed to solving the above-mentioned problems occurring in the prior art while keeping the advantages achieved by the prior art intact.

[0213] Embodiments of the present disclosure provide an object recognition apparatus and method thereof for recognizing a type of an external object using a reference grid map.

[0214] Another embodiment of the present disclosure provides an object recognition apparatus and method thereof for recognizing a type of an external object using a maximum value and a minimum value among a plurality of points obtained using a sensor (eg, a LiDAR sensor) to reduce a calculation amount of a processor.

[0215] Another embodiment of the present disclosure provides an object recognition device and method thereof for accurately recognizing a type of an external object to reduce errors in a vehicle control system including the object recognition device.

[0216] The technical problems to be solved by the present disclosure are not limited to the above-mentioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art to which the present disclosure belongs from the following description.

[0217] According to an embodiment of the present disclosure, an object recognition device may include a light detection and ranging device (LiDAR), a memory, and a processor. The processor may project a plurality of points indicating an external object obtained by LiDAR onto a first surface formed by a second axis and a third axis among a first axis, a second axis, and a third axis to obtain a first grid map including a first plurality of grids; may project a plurality of points indicating an external object obtained by LiDAR onto a second surface formed by a first axis and a third axis to obtain a second grid map including a second plurality of grids; and may identify the type of the external object based on matching at least one reference grid map in a reference grid map representing the appearance of a vehicle with at least one of the first grid map or the second grid map or any combination thereof, the reference grid map being stored in the memory.

[0218] In an embodiment, the processor may obtain at least one of the first grid map or the second grid map, or any combination thereof, based on the external object being identified as a moving state among a stationary state and a moving state.

[0219] In an embodiment, the processor can identify at least one point (e.g., all or some) among the multiple points included in the first multiple points in the first multiple grids, and can obtain a first grid map corresponding to the back surface of the external object based on the first grid, the first grid including all or some of the first multiple points included in the first multiple grids.

[0220] In an embodiment, the processor can identify a first maximum value, the first maximum value including a maximum value of a first plurality of points included in each of a plurality of columns, the plurality of columns being in a plurality of rows and a plurality of columns in a first grid map, can identify a first minimum value, the first minimum value including a minimum value of the first plurality of points included in each of the plurality of columns, and can obtain a first grid map indicating a rear contour of an external object based on the first maximum value and the first minimum value.

[0221] In an embodiment, the processor can identify a first upper grid corresponding to a first maximum value, a first lower grid corresponding to a first minimum value, and a first intermediate grid included between the first upper grid and the first lower grid; can identify a first reference grid included in a first reference grid map representing a rear surface of a vehicle in the reference grid map; can determine whether to perform a first upper match between the first upper grid and the first reference grid, can determine whether to perform a first lower match between the first lower grid and the first reference grid, can determine whether to perform a first intermediate match between the first intermediate grid and the first reference grid, and can identify the type of the external object based on the number of matches and the number of mismatches calculated by whether to perform the first upper match, whether to perform the first lower match, and whether to perform the first intermediate match.

[0222] In an embodiment, the processor may identify all or some of the plurality of points included in the second plurality of points in the second plurality of grids, and may obtain a second grid map corresponding to the side surface of the external object based on the second grid including all or some of the second plurality of points in the second plurality of grids.

[0223] In an embodiment, the processor can identify a second maximum value, which includes a maximum value of a second plurality of points included in each column of a plurality of rows and a plurality of columns included in the second grid map; can identify a second minimum value of the minimum value of the second plurality of points included in each column of the plurality of columns; and can obtain a second grid map indicating the rear contour of the external object based on the second maximum value and the second minimum value.

[0224] In an embodiment, the processor can identify a second upper grid corresponding to a second maximum value, a second lower grid corresponding to a second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; can identify a second reference grid included in a second grid map representing a side surface of the vehicle in the reference grid map; can determine whether to perform a second upper match between the second upper grid and the second reference grid; can determine whether to perform a second lower match between the second lower grid and the second reference grid; can determine whether to perform a second intermediate match between the second intermediate grid and the second reference grid; and can identify the type of the external object based on the number of matches and the number of mismatches calculated by whether to perform the second upper match, whether to perform the second lower match, and whether to perform the second intermediate match.

[0225] In an embodiment, the processor may assign a first score calculated based on the first grid map and the first reference grid map included in the reference grid map to the first grid map; may assign a second score calculated based on the second grid map and the second reference grid map included in the reference grid map to the second grid map; and may identify the type of the external object based on at least one of the first score or the second score or any combination thereof.

[0226] In an embodiment, if an external object is identified in the same lane as a main vehicle including an object recognition device, the processor may assign a first score to a first grid map; based on the first score being greater than a first reference score, the type of the external object may be identified; the first score may be assigned to the first grid map; if the external object is identified in a different lane from the main vehicle including the object recognition device, a second score may be assigned to a second grid map; if the external object is identified in a different lane from the main vehicle including the object recognition device, and based on the total score obtained by adding the first score to the second score being greater than a second reference score that is greater than the first reference score, the type of the external object may be identified or the type of the external object may be temporarily stored.

[0227] According to another embodiment of the present disclosure, an object recognition method may include: projecting multiple points indicating an external object obtained by LiDAR onto a first surface formed by a second axis and a third axis among a first axis, a second axis and a third axis to obtain a first grid map including a first plurality of grids; projecting multiple points indicating an external object obtained by LiDAR onto a second surface formed by a first axis and a third axis to obtain a second grid map including a second plurality of grids; and identifying the type of the external object based on matching at least one reference grid map among reference grid maps representing the appearance of a vehicle with at least one of the first grid map or the second grid map or any combination thereof, the reference grid maps being stored in a memory.

[0228] The object recognition method according to the embodiment may further include: obtaining at least one of the first grid map or the second grid map or any combination thereof based on the external object being recognized as a moving state among a stationary state and a moving state.

[0229] The object recognition method according to an embodiment may further include: identifying at least one point (for example, all or some) among the plurality of points included in a first plurality of points in a first plurality of grids, and obtaining a first grid map corresponding to a back surface of an external object based on a first grid including all or some of the first plurality of points in the first plurality of grids.

[0230] According to the object recognition method of the embodiment, the method may further include: obtaining a first maximum value of the maximum value of a first plurality of points included in each of a plurality of columns among a plurality of rows and a plurality of columns included in a first grid map; obtaining a first minimum value of the minimum value of the first plurality of points included in each of the plurality of columns; and obtaining a first grid map indicating the rear contour of an external object based on the first maximum value and the first minimum value.

[0231] The object recognition method according to the embodiment may further include: identifying a first upper grid corresponding to a first maximum value, a first lower grid corresponding to a first minimum value, and a first intermediate grid included between the first upper grid and the first lower grid; identifying a first reference grid included in a first reference grid map representing a rear surface of a vehicle in a reference grid map; determining whether to perform a first upper match between the first upper grid and the first reference grid; determining whether to perform a first lower match between the first lower grid and the first reference grid; determining whether to perform a first intermediate match between the first intermediate grid and the first reference grid; and identifying the type of the external object based on the number of matches and the number of mismatches calculated based on whether the first upper match is performed, whether the first lower match is performed, and whether the first intermediate match is performed.

[0232] The object recognition method according to an embodiment may further include: identifying at least one point (for example, all or some) among the multiple points included in a second multiple points in a second multiple grids, and obtaining a second grid map corresponding to the side surface of the external object based on a second grid in the second multiple grids including all or some of the second multiple points.

[0233] According to an embodiment, the object recognition method may further include: identifying a second maximum value and a second minimum value in a plurality of columns among a plurality of rows and a plurality of columns included in a second grid map (wherein some of the plurality of points are identified); and obtaining a second grid map representing a side profile of an external object based on the second maximum value and the second minimum value.

[0234] The object recognition method according to the embodiment may further include: identifying a second upper grid corresponding to a second maximum value, a second lower grid corresponding to a second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; identifying a second reference grid included in a second grid map representing a side surface of a vehicle in a reference grid map; determining whether to perform a second upper match between the second upper grid and the second reference grid; determining whether to perform a second lower match between the second lower grid and the second reference grid; determining whether to perform a second intermediate match between the second intermediate grid and the second reference grid; and identifying the type of the external object based on the number of matches and the number of mismatches calculated based on whether to perform the second upper match, whether to perform the second lower match, and whether to perform the second intermediate match.

[0235] The object recognition method according to the embodiment may further include: assigning a first score calculated based on the first grid map and the first reference grid map included in the reference grid map to the first grid map; assigning a second score calculated based on the second grid map and the second reference grid map included in the reference grid map to the second grid map; and identifying the type of the external object based on at least one of the first score or the second score or any combination thereof.

[0236] The object recognition method according to the embodiment may further include: if the external object is recognized in the same lane as the main vehicle including the object recognition device, assigning a first score to the first grid; identifying the type of the external object based on the first score being greater than a first reference score; if the external object is recognized in a lane different from that of the main vehicle including the object recognition device, assigning the first score to the first grid map; if the external object is recognized in a lane different from that of the main vehicle including the object recognition device, assigning a second score to the second grid map; and identifying the type of the external object or temporarily storing the type of the external object based on a total score obtained by adding the first score to the second score being greater than a second reference score greater than the first reference score.

[0237] The present technology can use a reference grid map to identify the type of external objects.

[0238] Furthermore, the present technology can identify the type of an external object using the maximum and minimum values ​​of a plurality of points obtained using LiDAR, thereby reducing the amount of calculation of the processor.

[0239] Furthermore, the present technology can accurately identify the type of external objects, thereby reducing errors in a vehicle control system including an object recognition device.

[0240] Furthermore, various effects directly or indirectly determined by the present disclosure can be provided.

[0241] Although the present disclosure has been described above with reference to the embodiments and the accompanying drawings, the present disclosure is not limited thereto, but various modifications and changes may be made by those skilled in the art without departing from the spirit and scope of the present disclosure as claimed in the appended claims.

[0242] Therefore, the embodiments of the present disclosure are not intended to limit the technical spirit of the present disclosure, but are provided for illustrative purposes only. The scope of the present disclosure should be interpreted based on the appended claims, and all technical concepts within the scope equivalent to the claims should be included in the scope of the present disclosure.

Claims

1. An object recognition device, comprising: sensor; a memory configured to store a reference grid map representing an appearance of a vehicle; as well as The processor is configured as: Projecting a plurality of points indicative of external objects onto the first surface to obtain a first mesh map comprising a first plurality of meshes, wherein: obtaining the external object based on sensing information obtained by the sensor, and The first surface is formed by the second axis and the third axis among a first axis, a second axis and a third axis, wherein the first axis is parallel to a traveling direction of the vehicle and perpendicular to the second axis, and the third axis is perpendicular to the first axis and the second axis; projecting the plurality of points indicative of the external object onto a second surface to obtain a second mesh map comprising a second plurality of meshes, wherein the second surface is formed by the first axis and the third axis; determining a type of the external object based on at least one of the reference grid maps and at least one of the first grid map and the second grid map; and A signal indicating the type of the external object is output.

2. The object recognition device according to claim 1, wherein: The processor is configured to: Based on the external object being determined to be in a moving state, at least one of the first grid map or the second grid map is obtained.

3. The object recognition device according to claim 1, wherein: The processor is configured to: determining at least one of a first plurality of points in the plurality of points, wherein the first plurality of points are included in the first plurality of grids; and The first mesh map is obtained based on a first mesh including at least one of the first plurality of points, wherein the first mesh map corresponds to a back surface of the external object.

4. The object recognition device according to claim 3, wherein: The processor is configured to: In the first grid map including a plurality of rows and a plurality of columns, determining a first maximum value including a maximum value of the first plurality of points included in each column of the plurality of columns; determining a first minimum value comprising minimum values ​​of the first plurality of points; as well as Based on the first maximum value and the first minimum value, the first grid map representing the back contour of the external object is obtained.

5. The object recognition device according to claim 4, wherein: The processor is configured to: Determine a first upper grid corresponding to the first maximum value, a first lower grid corresponding to the first minimum value, and a first intermediate grid included between the first upper grid and the first lower grid; determining a first reference grid included in a first reference grid map among the reference grid maps, wherein the first reference grid map represents a rear surface of the vehicle; determining whether to perform a first upper matching between the first upper grid and the first reference grid; determining whether to perform a first lower matching between the first lower grid and the first reference grid; determining whether to perform a first intermediate matching between the first intermediate grid and the first reference grid; and The type of the external object is determined based on determinations of whether the first upper matching is performed, whether the first lower matching is performed, and whether the first intermediate matching is performed.

6. The object recognition device according to claim 1, wherein: The processor is configured to: determining at least one of a second plurality of points in the plurality of points, wherein the second plurality of points are included in the second plurality of grids; and The second mesh map is obtained based on a second mesh including at least one of the second plurality of points, wherein the second mesh map corresponds to a side surface of the external object.

7. The object recognition device according to claim 6, wherein: The processor is configured to: In the second grid map including a plurality of rows and a plurality of columns, determining a second maximum value including a maximum value of the second plurality of points included in each column of the plurality of columns; determining a second minimum value comprising the minimum value of the second plurality of points; as well as The second grid map is obtained based on the second maximum value and the second minimum value, wherein the second grid map represents a side profile of the external object.

8. The object recognition device according to claim 7, wherein: The processor is configured to: determining a second upper grid corresponding to the second maximum value, a second lower grid corresponding to the second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; determining a second reference grid included in a second reference grid map in the reference grid map, wherein the second reference grid map represents a side surface of the vehicle; determining whether to perform a second upper matching between the second upper grid and the second reference grid; determining whether to perform a second lower matching between the second lower grid and the second reference grid; determining whether to perform a second intermediate matching between the second intermediate grid and the second reference grid; and The type of the external object is determined based on a determination of whether the second upper match is performed, whether the second lower match is performed, and whether the second middle match is performed.

9. The object recognition device according to claim 1, wherein: The processor is configured to: assigning a first score to the first grid map, wherein the first score is determined based on the first grid map and a first reference grid map included in the reference grid map; assigning a second score to the second grid map, wherein the second score is determined based on the second grid map and a second reference grid map included in the reference grid map; and A type of the external object is determined based on at least one of the first score and the second score.

10. The object recognition device according to claim 9, wherein: The processor is configured to: assigning the first score to the first grid map based on detecting the external object in the same lane as a host vehicle including the apparatus; determining a type of the external object based on the first score being greater than a first reference score; assigning the first score to the first grid map based on detecting the external object in a different lane than the host vehicle including the apparatus; assigning the second score to the second grid map based on detecting the external object in a different lane than a host vehicle including the apparatus; as well as The type of the external object is determined or information indicating the type of the external object is temporarily stored based on a total score obtained by adding the first score and the second score being greater than a second reference score greater than the first reference score.

11. An object recognition method, the object recognition method being executed by a processor and comprising: Projecting a plurality of points indicative of external objects onto the first surface to obtain a first mesh map comprising a first plurality of meshes, wherein: obtaining the external object based on sensing information obtained by the sensor, and The first surface is formed by a second axis and a third axis among a first axis, the second axis and the third axis, wherein the first axis is parallel to a traveling direction of the vehicle and perpendicular to the second axis, and the third axis is perpendicular to the first axis and the second axis; projecting the plurality of points indicative of the external object onto a second surface to obtain a second mesh map comprising a second plurality of meshes, wherein the second surface is formed by the first axis and the third axis; determining a type of the external object based on at least one of reference grid maps stored in a memory and at least one of the first grid map and the second grid map, wherein the reference grid map represents an appearance of the vehicle; and A signal indicating the type of the external object is output.

12. The object recognition method according to claim 11, further comprising: Based on the external object being determined to be in a moving state, at least one of the first grid map and the second grid map is obtained.

13. The object recognition method according to claim 11, further comprising: determining at least one of a first plurality of points in the plurality of points, wherein the first plurality of points are included in the first plurality of grids; as well as The first mesh map is obtained based on a first mesh including at least one of the first plurality of points, wherein the first mesh map corresponds to a back surface of the external object.

14. The object recognition method according to claim 13, further comprising: In the first grid map including a plurality of rows and a plurality of columns, obtaining a first maximum value including a maximum value of the first plurality of points included in each column of the plurality of columns; obtaining a first minimum value including minimum values ​​of the first plurality of points; as well as Based on the first maximum value and the first minimum value, the first grid map representing the back contour of the external object is obtained.

15. The object recognition method according to claim 14, further comprising: Determine a first upper grid corresponding to the first maximum value, a first lower grid corresponding to the first minimum value, and a first intermediate grid included between the first upper grid and the first lower grid; determining a first reference grid included in a first reference grid map among the reference grid maps, wherein the first reference grid map represents a rear surface of the vehicle; determining whether to perform a first upper matching between the first upper grid and the first reference grid; determining whether to perform a first lower matching between the first lower grid and the first reference grid; determining whether to perform a first intermediate matching between the first intermediate grid and the first reference grid; and The type of the external object is determined based on determinations of whether the first upper matching is performed, whether the first lower matching is performed, and whether the first intermediate matching is performed.

16. The object recognition method according to claim 11, further comprising: determining at least one of a second plurality of points in the plurality of points, wherein the second plurality of points are included in the second plurality of grids; as well as The second mesh map is obtained based on a second mesh including at least one of the second plurality of points, wherein the second mesh map corresponds to a side surface of the external object.

17. The object recognition method according to claim 16, further comprising: In the second grid map including a plurality of rows and a plurality of columns, determining a second maximum value including a maximum value of the second plurality of points included in each column of the plurality of columns; determining a second minimum value comprising the minimum value of the second plurality of points; as well as The second grid map is obtained based on the second maximum value and the second minimum value, wherein the second grid map represents a side profile of the external object.

18. The object recognition method according to claim 17, further comprising: determining a second upper grid corresponding to the second maximum value, a second lower grid corresponding to the second minimum value, and a second intermediate grid included between the second upper grid and the second lower grid; determining a second reference grid included in a second reference grid map in the reference grid map, wherein the second reference grid map represents a side surface of the vehicle; determining whether to perform a second upper matching between the second upper grid and the second reference grid; determining whether to perform a second lower matching between the second lower grid and the second reference grid; determining whether to perform a second intermediate matching between the second intermediate grid and the second reference grid; and The type of the external object is determined based on a determination of whether the second upper matching is performed, whether the second lower matching is performed, and whether the second middle matching is performed.

19. The object recognition method according to claim 11, further comprising: assigning a first score to the first grid map, wherein the first score is determined based on the first grid map and a first reference grid map included in the reference grid map; assigning a second score to the second grid map, wherein the second score is determined based on the second grid map and a second reference grid map included in the reference grid map; and A type of the external object is determined based on at least one of the first score and the second score.

20. The object recognition method according to claim 19, further comprising: assigning the first score to the first grid map based on detecting the external object in the same lane as a host vehicle including the processor, and determining a type of the external object based on the first score being greater than a first reference score; or Based on detecting the external object in a lane different from that of the host vehicle including the processor, assigning the first score to the first grid map and assigning the second score to the second grid map; and determining the type of the external object or temporarily storing the type of the external object based on a total score obtained by adding the first score to the second score being greater than a second reference score greater than the first reference score.

Images