DEVICE AND METHOD FOR DETECTING OBJECTS

Abstract

The present disclosure relates to an apparatus and method for object detection, and the object detection apparatus includes a sensor and a processor. The processor may determine longitudinal position information and lateral position information based on sensing information from the sensor, lane information, whether at least one other object different from the object and detected as a moving object or a stationary object is located within the lane in which the object is located, assign a reliability value to the object, wherein, based on the at least one other object being located within the lane in which the object is located, a first value corresponding to the reliability value for the object or a second value less than the first value is used as the reliability value, and determine the object as a moving object or a stationary object.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to an apparatus and method for object recognition, and more particularly to a technique for identifying or recognizing properties of an object based on a contour point detected by one or more sensors, such as a Light Detection and Ranging (LIDAR). BACKGROUND

[0002] Environment sensing and obstacle discrimination technology is required to enable a vehicle to adjust its course and avoid obstacles without driver intervention.

[0003] A vehicle can obtain data indicating the position of an object near the vehicle through a sensor such as a LIDAR device (hereinafter referred to as "LIDAR"). A distance from a LIDAR to an object can be obtained by an interval between the time a laser is transmitted from the LIDAR and the time the laser reflected from the object is received. A vehicle is capable of detecting the position of a point included in the object in a space where the vehicle is located based on the angle of the transmitted laser and the distance to the object.

[0004] The autonomous vehicle can process the data collected by the LIDAR to determine information about the environment and surrounding objects. However, due to the limited memory and time available for data processing, there is a growing need for technologies that can efficiently extract information about the environment and surrounding objects from the data collected by the LIDAR. SUMMARY

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

[0006] An embodiment of the present disclosure provides an apparatus and method for object recognition to identify whether an object is a moving object or a stationary object (e.g., an object that is not currently in a state of motion but is capable of being in a state of motion, an object that is not capable of not being in a state of motion, etc.).

[0007] An embodiment of the present disclosure provides an apparatus and method for object detection to detect whether an object having a part occluded or being in a stationary state is a moving object or a stationary object.

[0008] An embodiment of the present disclosure provides an apparatus and method for object recognition for improving the accuracy of determining whether an object is a moving object or a stationary object.

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

[0010] A device may comprise: a sensor; and a processor, wherein the processor is configured to: determine, based on sensing information of the sensor, longitudinal position information representing a distance in a line direction from a vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction; determine, based on a determination that the longitudinal position information falls within a range according to a threshold distance, lane information representing a lane in which the object is located, wherein the lane information is determined based on at least one of: the lateral position information or positions of lines located on both sides of the vehicle;to determine whether at least one other object, different from the object and detected as a moving object or a stationary object, is located within the lane in which the object is located; assigning a confidence value to the object, wherein a first value corresponds to the confidence value for the object based on the at least one other object being within the lane in which the object is located, or a second value, less than the first value, corresponds to the confidence value for the object based on the at least one other object not being within the lane in which the object is located; based on the confidence value, determining the object as a moving object or as a stationary object;and output a signal indicating that the object is a moving object or a signal indicating that the object is a stationary object;

[0011] The processor may be configured to: Detect curvature information indicating a curvature of a lane in which the vehicle is located, and detect the positions of the lines based on at least one of: the longitudinal position information, the curvature information, a width of a particular lane, or a number of particular lanes.

[0012] The processor may be configured to: detect, based on the lateral position information, that the object is located between a position of a first line and a position of a second line adjacent to the first line, and wherein the lane information is configured to indicate a lane located between the first line and the second line.

[0013] A threshold distance of longitudinal position information detected by the vehicle traveling at a first speed may be configured to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling at a second speed that is higher than the first speed.

[0014] A threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a first curvature may be configured to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a second curvature that is greater than a first curvature.

[0015] The processor may be configured to: Determine curvature information indicating a curvature of a lane in which the vehicle is located; and determine the longitudinal position information and the lateral position information based on the curvature information being greater than a certain curvature value.

[0016] A third value corresponds to the confidence value for the object when a first number of other objects detected as moving objects or objects capable of being in a state of movement are detected within the lane in which the object is located; and a fourth value, greater than the third value, corresponds to the confidence value for the object when a second number of other objects, greater than the first number, are detected within the lane in which the object is located.

[0017] The processor may be configured to: change the reliability value from the second value to the first value based on the lane information of the object being changed as the object moves from a first lane in which no other object is located to a second lane in which the other object is located.

[0018] The processor may be configured to: maintain the confidence value even if at least one other object moves into the lane in which the object is located before the object is detected as a moving object or before the object is detected as a stationary object.

[0019] The processor may be configured to: assign to the object an identifier indicating that the object is a moving object or a stationary object capable of being in a state of motion based on a determination that a score value representing a probability that the object detected based on the confidence value is a moving object or a stationary object capable of being in a state of motion is greater than a score value representing a probability that the object is a stationary object incapable of being in a state of motion.

[0020] A method may include: determining, by a processor and based on sensing information from a sensor, longitudinal position information representing a distance in a line direction from a vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction; determining, based on a determination that the longitudinal position information falls within a range according to a threshold distance, lane information representing a lane in which the object is located, wherein the lane information is determined based on at least one of: the lateral position information or positions of lines located on both sides of the vehicle;Determining whether at least one other object that is different from the object and that is detected as a moving object or a stationary object is located within the lane in which the object is located; Assigning a confidence value to the object, wherein a first value corresponds to the confidence value for the object based on the at least one other object being within the lane in which the object is located, or a second value that is less than the first value corresponds to the confidence value for the object based on the at least one other object not being within the lane in which the object is located; Determining, based on the confidence value, the object as a moving object or as a stationary object;and outputting a signal indicating that the object is a moving object or a signal indicating that the object is a stationary object;

[0021] The method may further include one or more of the operations and functions or features described herein.

[0022] These and other features and advantages are described in more detail below. The features of the present disclosure briefly summarized above are merely exemplary embodiments of the detailed description of the present disclosure described below and do not limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings: Fig. 1 is a block diagram showing an object detection device; Fig. 2 shows a table showing information required to recognize whether an object in an object recognition device or an object recognition method is a moving object, an object capable of being in a moving state, or an object incapable of being in a moving state; Fig. 3 shows a range in which the weights of the reliability values ​​vary according to information in an object recognition device or an object recognition method; Fig. 4 shows an example of detecting reliability according to information performed on a partially occluded object in an object detection apparatus or method; Fig. 5 is a flowchart showing an operation of an object recognition device for classifying objects based on reliability according to information in the object recognition device or an object recognition method; Fig. 6 is a flowchart showing an operation of an object recognition device for detecting a threshold distance in the object recognition device or an object recognition method; Fig. 7 shows an example of lane information where an object is located in an object recognition device or method; Fig. 8 shows an example of a change in a reliability value according to information that changes when a host vehicle changes a lane in an object detection device or method; Fig. 9 shows an example of a threshold distance that changes according to the speed of a host vehicle and curvature information in an object detection device or method; Fig. 10 shows an example of an object whose reliability is detected within a threshold distance in an object detection apparatus or method; Fig. 11 shows an example of a change in a reliability value according to information that changes depending on the positions of other objects in an object recognition device or an object recognition method; and Fig. 12 shows a computer system relating to an object recognition device or method. DETAILED DESCRIPTION

[0024] Various embodiments of the present disclosure will be described in detail below with reference to the exemplary drawings. When adding reference numerals to the components of each drawing, it should be noted that the identical or comparable component will be designated by the identical reference numeral even if they are shown in different drawings. Furthermore, in describing various features of the present disclosure, a detailed description of well-known features or functions may be omitted in order not to unnecessarily obscure the essence of the present disclosure.

[0025] In describing the components of the embodiment according to the present disclosure, terms such as "first," "second," "A," "B," "(a)," "(b)," and the like may be used. These terms are merely intended to distinguish one component from another, and the terms do not limit the nature, sequence, or order of the individual components. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure relates.Such terms, as defined in a commonly used dictionary, shall be construed to have a meaning consistent with the contextual meaning in the relevant field of prior art and shall not be construed to have an ideal or overly formal meaning unless clearly defined as such in the present application.

[0026] In addition, the terms “unit”, “device”, “element”, “body” or the like used below may indicate at least one shape structure or may indicate a unit for processing a function.

[0027] Furthermore, in the present disclosure, the terms "greater than" or "less than" may be used to indicate whether a particular condition is met or satisfied, but they are used only to provide examples and do not exclude "greater than or equal to" or "less than or equal to." A condition indicating "greater than or equal to" may be replaced with "greater than," a condition indicating "less than or equal to" may be replaced with "less than," and a condition indicating "greater than or equal to and less than" may be replaced with "greater than and less than or equal to." Furthermore, "A" to "B" means at least one of A (inclusive) to B (inclusive).

[0028] Hereinafter, examples of the present disclosure are described with reference to the Fig. 1 to 12 described in detail.

[0029] Fig. 1 shows a block diagram showing an object detection device.

[0030] With reference to Fig. 1, an object detection device 101 may be implemented inside (or outside) a vehicle. In this case, the object detection device 101 may be formed integrally with internal control units of the vehicle or may be implemented as a separate device and connected to the control units of the vehicle by separate connection means.

[0031] With reference to Fig. 1, the object detection device 101 may comprise a sensor (e.g., a LIDAR 103) and a processor 105.

[0032] The processor 105 of the object detection device 101 can obtain position information of points of an object near a vehicle (e.g., a vehicle including the object detection device 101) via the LIDAR 103. The processor 105 of the object detection device 101 can detect a plurality of points (e.g., a point cloud) representing the object via the LIDAR 103. The processor 105 of the object detection device 101 can detect contour points among points included in the plurality of points (e.g., a point cloud).

[0033] The processor 105 of the object recognition device 101 can recognize position information of a point corresponding to an object based on contour points representing the object. For example, the point corresponding to the object may include a center point of the rearmost line segment in the moving direction of the object among line segments forming an object box including contour points representing the object. The center point may be referred to as a tracking point, but is not limited to this.

[0034] The processor 105 of the object detection device 101 can detect longitudinal position information indicating a distance from the host vehicle to an object in the line direction. The processor 105 of the object detection device 101 can detect lateral position information indicating a distance from the host vehicle to the object in a direction perpendicular to the line direction.

[0035] The processor 105 of the object recognition device 101 can recognize curvature information indicating the curvature of a lane based on a speed of the host vehicle and a yaw rate of the host vehicle.

[0036] The processor 105 of the object recognition device 101 can recognize a threshold distance based on the curvature information. The threshold distance may refer to a distance including an object that is capable of guaranteeing a certain value or more of accuracy for an object recognition method. The processor 105 of the object recognition device 101 can recognize the threshold distance based on at least one of the curvature information, the speed information of the host vehicle, or any combination thereof.

[0037] The processor 105 of the object recognition device 101 can detect whether the longitudinal position information of the object falls within a range according to the threshold distance. If the longitudinal position information of the object falls within the range according to the threshold distance, the object recognition device 101 can determine whether the object is a moving object or a stationary object of a first type (e.g., an object capable of being in a state of motion). The moving object or the stationary object of a first type may include vehicles (e.g., two-wheeled vehicles, four-wheeled vehicles). A stationary object of a second type (e.g., an object incapable of being in a state of motion) may include structures (e.g., traffic signs, guardrails).

[0038] The processor 105 of the object recognition device 101 can recognize the positions of lines located on both sides of the host vehicle based on at least one of the curvature information, the longitudinal position information of the object, the width of a particular lane (e.g., about 3.5 m or about 3.7 m), or the number of particular lanes (e.g., 7), or any combination thereof. The recognition of the positions of lines will be described further below with reference to Fig. 7 described.

[0039] The processor 105 of the object recognition device 101 can recognize the longitudinal position information of the object and lane information indicating a lane in which the object is located according to the positions of the lines. The recognition of the lane information will be described below with reference to Fig. 7 described.

[0040] Based on the lane information, the processor 105 of the object detection device 101 may detect that at least one other object is located within the lane in which the object is located. If at least two other objects are detected as moving objects or objects that may be in a state of movement, the processor 105 of the object detection device 101 may assign a first value (e.g., 1) as a reliability value according to information for other objects within the lane.

[0041] When an object other than a moving object or an object that may be in a state of motion is detected, the processor 105 of the object detection device 101 may assign a second value (e.g., 0.8) as a confidence value according to information for other objects within the object's lane.

[0042] If it is determined based on lane information that at least one other object is not in the lane in which the object is located, the processor 105 of the object recognition device 101 may assign a third value (e.g., 0) smaller than the first value as a confidence value according to information for other objects within the object's lane.

[0043] The processor 105 of the object recognition device 101 may calculate a score value based on the object's reliability value, which indicates a probability that the object is a moving object or an object that may be in a moving state. The reliability value may be recognized based on the information for other objects within the lane. The information for other objects within the lane may be referred to as a lane dynamic object variable, but embodiments of the present disclosure are not limited thereto. The recognition of the score value based on the reliability value according to the object's information will be described below with reference to Fig. 2 described.

[0044] The processor 105 of the object recognition device 101 can recognize that an object is a moving object or an object capable of being in a moving state based on the score value indicating the probability that an object is a moving object or an object capable of being in a moving state.

[0045] Based on determining that the object is a moving object or an object that may be in a state of motion, the processor 105 of the object recognition device 101 may assign an identifier to the object indicating that the object is a moving object or an object that may be in a state of motion. The identifier may be referred to as, but is not limited to, a flag.

[0046] Fig. 2 shows a table showing information required to detect whether an object in an object detection device or method is a moving object, an object that can be in a moving state, or an object that cannot be in a moving state.

[0047] With reference to Fig. 2, Table 201 may represent types of information for calculating a score to determine whether an object is a moving object or an object that may be in a state of motion. An immobility score 203 may represent a score to determine whether an object is an object that may not be in a state of motion. The immobility score 203 may be determined based on information such as out-of-lane information 211, box size information 213, and box matching information 215. A mobility score 205 may represent a score to determine whether an object is a moving object or an object that may be in a state of motion.The mobility score 205 may be recognized based on information such as lane information 217, tracking information 219, information 221 for other objects within the lane, speed information 223, contour point distribution information 225, and boundary object information 227.

[0048] The out-of-lane information 211 for detecting the immobility score 203 may represent a reliability assigned based on whether an object is detected outside a lane. The box size information 213 for detecting the immobility score 203 may represent a reliability assigned based on whether the size of an object box is greater than or equal to a reference size. The box matching information 215 for detecting the immobility score 203 may represent a reliability assigned based on the distribution of the contour points and the degree of matching of the object box.

[0049] The in-lane information 217 may represent a confidence assigned based on whether an object is detected within a lane. The tracking information 219 may represent a confidence assigned based on whether an object is moving. The speed information 223 may represent a confidence assigned based on the speed of an object. The boundary object information 227 may represent a confidence assigned based on whether an object is detected without being occluded at the boundary of a field of view.

[0050] The immobility score 203 can be recognized by the sum of values ​​obtained by multiplying reliabilities (e.g., reliability values ​​or scores) represented by partial information by a weight. For example, the immobility score 203 can be recognized by the sum of a value obtained by multiplying the reliability value according to the out-of-lane information 211 by a weight corresponding to the out-of-lane information 211 (e.g., weight S1 ), a value obtained by multiplying the reliability value according to the box size information 213 by a weighting corresponding to the box size information (e.g. weighting S2), a value obtained by multiplying the reliability value according to the box-matching information 215 by a weighting corresponding to the box-matching information (e.g., weighting S3 ), or at least any combination thereof. However, embodiments of the present disclosure are not limited thereto. According to one example, the immobility score 203 may be recognized by adding not only a value obtained by multiplying a value (e.g., a score) corresponding to a piece of information listed in the table 201 by a weight, but also a value obtained by multiplying a value (e.g., a score) corresponding to a piece of information not listed in the table 201 by the weight.

[0051] The mobility score 205 may be recognized by the sum of values ​​obtained by multiplying the reliability values ​​(or scores) represented by partial information indicating the reliabilities by a weight. For example, the mobility score 205 may be recognized by the sum of at least one of a value obtained by multiplying the reliability value according to the in-lane information 217 by a weight corresponding to the in-lane information 217 (e.g., weight D1 ), a value obtained by multiplying the reliability value according to the tracking information 219 by a weighting corresponding to the tracking information 219 (e.g. weight D2), a value obtained by multiplying the reliability value according to the information 221 for other objects within the lane by a weighting corresponding to the information 221 for other objects within the lane (e.g. weight D3 ), a value obtained by multiplying the reliability value according to the speed information 223 by a weighting corresponding to the speed information 223 (e.g. weighting D4 ), a value obtained by multiplying the reliability value according to the contour point distribution information 225 by a weighting corresponding to the contour point distribution information 225 (e.g., weighting D5 ), a value obtained by multiplying the reliability value according to the boundary object information 227 by a weighting corresponding to the boundary object information 227 (e.g. weighting D6), or any combination thereof. However, embodiments of the present disclosure are not limited thereto. The mobility score 205 can be recognized by adding not only a value obtained by multiplying a value (or score) corresponding to a piece of partial information listed in the table 201 by a weight, but also a value obtained by multiplying a value (or score) corresponding to a piece of partial information not listed in the table 201 by the weight.

[0052] If the mobility score 205 for a particular object is higher than the immobility score 203 for the particular object, the processor of the object recognition device may recognize that the particular object is a moving object or an object that may be in a state of motion. If the immobility score 203 for a particular object is higher than the mobility score 205 for the particular object, the processor of the object recognition device may recognize that the particular object is an object that may not be in a state of motion.

[0053] According to the present disclosure, a device (e.g., the object recognition device 101) may recognize the reliability value represented by the information 221 for other objects within the lane. A method for recognizing a reliability value represented by the information 221 for other objects within the lane is described below with reference to Fig. 5 described.

[0054] Fig. 3 shows a range in which the weights of reliability values ​​vary according to information in an object recognition device or an object recognition method.

[0055] With reference to Fig. 3, a frame 301 may represent a first region 305, a second region 307, and a third region 309 separated according to a distance from a host vehicle 303 (e.g., a vehicle including the object detection device). The first region 305 may include a region within a field of view. The second region 307 may include a region for classifying objects of interest. The third region 309 may include regions other than the first region and the second region.

[0056] The first region 305 may be referred to as a field of view (FOV), but is not necessarily limited thereto. The second region 307 may be referred to as a class region of interest (Class ROI), but is not necessarily limited thereto. The third region may be referred to as a standard region, but is not necessarily limited thereto.

[0057] The processor of the object recognition device may assign different weights (e.g., weights in Fig. 2). This is because information of high importance may vary depending on the position of an object. For example, the processor of the object recognition device may set a weight of the contour point distribution information 225 to a value greater than 0 only in the second region 307. For example, the processor of the object recognition device may set the weight of the boundary object information 227 in the first region 305 higher than the weight of the boundary object information 227 in the second region 307 and the weight of the boundary object information 227 in the third region 309.

[0058] Fig. 4 shows an example of detecting a reliability value according to the information obtained for a partially occluded object in an object detection apparatus or method.

[0059] With reference to Fig. 4, as in screen 401, a first object 403 may be partially occluded by another object (e.g., obscured in a field of view). The processor of a host vehicle's object detection device 413 may capture a frame 411 that includes contour points obtained by a LIDAR. The frame 411 may include a first object 415 and a second object 417.

[0060] The processor of the object detection device may need to detect whether a partially occluded object, such as the first object 403 on the screen 401, is a moving object, an object that may be in a state of motion, or an object that may not be in a state of motion. This is because the motion of the host vehicle cannot be determined until it is detected that the first object 403 on the screen 401 is an object that may not be in a state of motion.

[0061] However, an existing object detection device may have difficulty detecting whether the partially occluded object 403 is a moving object, an object that may be in a state of motion, or an object that may not be in a state of motion. This is because the shape of the object is not fully detected.

[0062] The processor of the object detection device may use a confidence value included in the information for other objects within the lane to detect whether the partially occluded object 403 is a moving object, an object that may be in a moving state, or an object that may not be in a moving state.

[0063] The processor of the object detection device may detect in frame 411 that the second object 417 of frame 411 obscures the first object 415 of frame 411. If the second object 417 of frame 411 is a moving object or an object that may be in a moving state in the same lane as the first object 415 of frame 411, the processor of the object detection device may assign a confidence value to the first object 415 of frame 411 according to information for other objects within the lane, even though a portion of the first object 415 of frame 411 is obscured. The processor of the object detection device may detect whether the first object 415 of frame 411 is a moving object or an object that may be in a moving state based on the confidence value.

[0064] The reason for this is that a probability that the particular object is a moving object or an object that may be in a state of motion if another object that exists in the same lane as the lane of the particular object is a moving object or an object that may be in a state of motion is higher than a probability that the particular object is a moving object or an object that may be in a state of motion if a moving object or an object that may be in a state of motion does not exist in the same lane as the lane of the particular object.

[0065] Accordingly, one or more features of the present disclosure may be effective and useful for detecting whether a partially occluded object is a moving object or an object that may be in a state of motion.

[0066] Furthermore, one or more features of the present disclosure may be effective and useful for detecting whether an object in a stationary state (e.g., a preceding, stationary vehicle, a traffic sign) is an object that may be in a moving state (e.g., a vehicle).

[0067] The processor of the object recognition device may recognize in frame 411 that the first object 415 of frame 411 is in a stationary state. The processor of the object recognition device may assign a reliability value to the first object 415 of frame 411 according to information for other objects within the lane if the second object 417 of frame 411, which is in the same lane as the first object 415 of frame 411, is a moving object or an object that may be in a state of movement even though the first object 415 of frame 411 is in a stationary state. The processor of the object recognition device may recognize whether the first object 415 of frame 411 is an object that may be in a state of movement based on the reliability value.

[0068] The following describes a method for detecting whether an object is a moving object or an object that may be in a state of movement by assigning a reliability value according to another object within the lane (see Fig. 5 to 7).

[0069] Fig. 5 shows a flowchart of an operation of an object recognition device for classifying objects based on a reliability value according to information in the object recognition device or an object recognition method.

[0070] In the following, it is assumed that the object recognition device 101 of Fig. 1 the process of Fig. 5. In addition, the description of Fig. 5 Operations described as being performed by the device are understood to be controlled by the processor 105 of the object recognition device 101, but embodiments are not limited as such.

[0071] With reference to Fig. 5, in a first operation 501, the processor of an object detection device may detect curvature information, longitudinal position information, and lateral position information. The curvature information may indicate the curvature of a lane in which a host vehicle is located. The longitudinal position information may indicate a distance from the host vehicle to an object in a line direction. The lateral position information may indicate a distance from the host vehicle to an object in a direction perpendicular to the line direction.

[0072] The curvature information may be detected based on the yaw rate and speed of the host vehicle including the object detection device. The curvature information may be detected based on the radius of curvature.

[0073] The processor of the object recognition device may recognize the longitudinal position information and the lateral position information based on determining that the curvature information is greater than a certain curvature value (e.g., about 200 m). This is because if the curvature information is less than or equal to the certain curvature value, the accuracy of the lane information will be reduced. A situation in which the curvature information is less than or equal to the certain curvature value may include a situation in which the host vehicle is turning (e.g., making a U-turn).

[0074] In a second operation 503, the processor of the object detection device may detect lane information based on determining that the longitudinal position information is included within a range according to a threshold distance. The threshold distance may refer to a distance to an object capable of guaranteeing accuracy of a certain value or more for an object detection method. The lane information may indicate a lane in which an object is located.

[0075] The processor of the object detection device may detect the threshold distance based on at least one of the curvature information, the speed of the host vehicle, or any combination thereof.

[0076] For example, as the curvature of the lane in which the host vehicle is located increases, the threshold distance may decrease. This is because when the curvature of the lane is large, the error of the lane information for detecting whether there is another object within the lane, which is a moving object or an object that may be in a moving state, may increase.

[0077] For example, as the host vehicle's speed decreases, the threshold distance may decrease. This is because the need to detect whether the distant objects are moving objects or objects that may be in a state of motion when the host vehicle's speed is low may be less than the need to detect whether the distant objects are moving objects or objects that may be in a state of motion when the host vehicle's speed is high.

[0078] The processor of the object detection device can detect the positions of the lines based on at least one of the curvature information, the width of a particular lane, the number of particular lanes, or any combination thereof. In this case, the processor of the object detection device can detect the lateral positions of the lines in a longitudinal position where an object is located according to the longitudinal position information of the object. The processor of the object detection device can detect lane information about a lane where an object is located by comparing the lateral positions of the lines and the lateral position of the object.

[0079] The lane information may be detected based on at least one of the object's lateral position information or the positions of lines located on both sides of the host vehicle, or any combination thereof. For example, if the object's lateral position information is located between the position of a first line and the position of a second line, which is the next line after the first line, the processor of the object detection device may detect that the object is located in a lane between the first lane and the second lane.

[0080] In a third operation 505, the processor of the object detection device may detect, based on the lane information, whether at least one other object detected as a moving object or an object that may be in a state of motion is located within the lane. If at least one other object detected as a moving object or an object that may be in a state of motion is located within the lane, the processor of the object detection device may perform a fourth operation 507. If at least one other object detected as a moving object or an object that may be in a state of motion is not located within the lane, the processor of the object detection device may perform a fifth operation 509.

[0081] At least one other object detected as a moving object or an object that may be in a state of motion may be located within the lane in which the object is located. A probability that an object is a moving object or an object that may be in a state of motion when a moving object or an object that may be in a state of motion is in the same lane as the lane in which the object is located is higher than a probability that the particular object is a moving object or an object that may be in a state of motion when a moving object or an object that may be in a state of motion is not in the same lane as the lane in which the object is located.

[0082] In the fourth operation 507, the processor may assign a first value to the object recognition device as a reliability value of the object.

[0083] If another object, which is a moving object or an object that may be in a state of motion, is in the same lane as the object, the first value may include 0.8.

[0084] If two or more other objects, which are moving objects or objects that may be in a state of motion, are in the same lane as the object, the first value may include 1.

[0085] The more other objects are detected within the lane in which the object is located, the higher the object's confidence value can be assigned. For example, the processor of the object detection device can assign a third value as a confidence value when a first number of other objects, which are detected as moving objects or objects that may be in a state of movement, are detected in the lane in which the object is located, and assign a fourth value, greater than the third value, as a confidence value when a second number of other objects are detected within the lane in which the object is located, the second number being greater than the first number.

[0086] In the fifth operation 509, the processor of the object recognition device may assign a second value, which is smaller than the first value, as the object's reliability value. For example, the second value may include 0.

[0087] In a sixth operation 511, the processor of the object recognition device may recognize the object as a moving object or an object that may be in a moving state based on the reliability value.

[0088] As in Fig. 2, the processor of the object recognition device may determine a score value indicating a probability, based on the confidence value, that an object is a moving object or an object that may be in a state of motion. The processor of the object recognition device may recognize that a particular object is a moving object or an object that may be in a state of motion based on a score value indicating the probability that a particular object is a moving object or an object that may be in a state of motion being greater than a score value indicating the probability that the particular object is an object that may not be in a state of motion.

[0089] The processor of the object recognition device may assign an identifier to the particular object indicating that the particular object is a moving object or an object that may be in a state of motion based on the particular object being recognized as a moving object or an object that may be in a state of motion.

[0090] Fig. 6 shows a flowchart of an operation of an object recognition device for detecting a threshold distance in the object recognition device or an object recognition method.

[0091] In the following, it is assumed that the object recognition device 101 of Fig. 1 the process of Fig. 6. In addition, the description of Fig. 6 Operations described as being performed by the device are understood to be controlled by the processor 105 of the object recognition device 101.

[0092] One or more features of the present disclosure may be applied to objects whose longitudinal position information is within a threshold distance. An operation of detecting the threshold distance is described below.

[0093] With reference to Fig. 6, the processor of the object detection device may detect whether the speed of a host vehicle is less than approximately 10 km / h in a first operation 601. If the speed of the host vehicle is less than approximately 10 km / h, the processor of the object detection device may perform a second operation 605. If the speed of the host vehicle is greater than or equal to approximately 10 km / h, the processor of the object detection device may perform a third operation 603.

[0094] The threshold distance may be detected according to the speed of the host vehicle and the curvature of a lane in which the host vehicle is located. When the speed of the host vehicle is below a first reference speed value (e.g., about 10 km / h), the threshold distance may be smaller than the threshold distance when the speed of the host vehicle is greater than or equal to the first reference speed value. In other words, the threshold distance of the longitudinal position information detected by the host vehicle traveling at a first speed (e.g., a speed less than about 10 km / h) may be smaller than the threshold distance of the longitudinal position information detected by the host vehicle traveling at a second speed (e.g., a speed of about 10 km / h or more) that is higher than the first speed.This is because the need to detect whether the remote objects are moving objects or objects that may be in a state of motion when the speed of the host vehicle is less than the first reference value may be less than the need to detect whether the remote objects are moving objects or objects that may be in a state of motion when the speed of the host vehicle is greater than or equal to the first reference value.

[0095] In the second operation 605, the processor of the object detection device may set the threshold distance to approximately 20 m.

[0096] In the third operation 603, the object detection device processor may detect whether the speed of the host vehicle is less than approximately 20 km / h. If the speed of the host vehicle is less than approximately 20 km / h, the object detection device processor may perform a fourth operation 609. If the speed of the host vehicle is greater than or equal to approximately 20 km / h, the object detection device processor may perform a fifth operation 607. The object detection device processor may detect whether the speed of the host vehicle is greater than or equal to the first reference speed value (e.g., approximately 10 km / h) and less than a second reference speed value (e.g., approximately 20 km / h).

[0097] In the fourth operation 609, the processor of the object detection device may set the threshold distance to approximately 30 m. Since the speed of the host vehicle when performing the fourth operation 609 is greater than the speed of the host vehicle when performing the second operation 605, the threshold distance in the fourth operation 609 may be greater than the threshold distance in the second operation 605.

[0098] In the fifth operation 607, the processor of the object recognition device may detect whether a curvature is greater than approximately 10,000 m. If the curvature is greater than approximately 10,000 m, the processor of the object recognition device may perform a sixth operation 613. If the curvature is less than or equal to approximately 10,000 m, the processor of the object recognition device may perform a seventh operation 611.

[0099] The threshold distance may be smaller when the curvature of a lane in which the host vehicle is located is less than or equal to a first reference curvature value (e.g., about 10,000 m), than the threshold distance when the curvature of the lane is greater than the first reference curvature value. In other words, the threshold distance of the longitudinal position information detected by the host vehicle traveling in a lane with a first curvature (e.g., a curvature of about 10,000 m or less) may be smaller than the threshold distance of the longitudinal position information detected by the host vehicle traveling in a lane with a second curvature (e.g., a curvature of more than about 10,000 m) that is greater than the first curvature.The reason is that the accuracy of detecting whether a distant object is in the same lane as another object when the curvature of the lane is greater than the first reference curvature value may be greater than the accuracy of detecting when the curvature of the lane is smaller than the first reference curvature value.

[0100] In the sixth operation 613, the processor of the object detection device may set the threshold distance to approximately 80 m.

[0101] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is greater than a first reference curvature value (e.g., about 10,000 m) or less than the threshold distance if the curvature of the lane is less than or equal to the first reference curvature value.

[0102] In the seventh operation 611, the object recognition device processor may detect whether the curvature is greater than approximately 5000 m. If the curvature is greater than approximately 5000 m, the object recognition device processor may perform an eighth operation 617. If the curvature is less than or equal to approximately 5000 m, the object recognition device processor may perform a ninth operation 615.

[0103] The processor of the object detection device can detect whether the curvature of a lane is less than or equal to the first reference curvature value (e.g., about 10,000 m) and greater than a second reference curvature value (e.g., about 5,000 m).

[0104] In the eighth operation 617, the processor of the object detection device may set the threshold distance to approximately 70 m.

[0105] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is less than or equal to the first reference curvature value (e.g., about 10,000 m) and greater than the second reference curvature value (e.g., about 5,000 m).

[0106] In the ninth operation 615, the object recognition device processor may detect whether the curvature is greater than approximately 3000 m. If the curvature is greater than approximately 3000 m, the object recognition device processor may perform a tenth operation 621. If the curvature is less than or equal to approximately 3000 m, the object recognition device processor may perform an eleventh operation 619.

[0107] The processor of the object detection device can detect whether the curvature of a lane is less than or equal to the second reference curvature value (e.g., about 5000 m) and greater than a third reference curvature value (e.g., about 3000 m).

[0108] In the tenth operation 621, the processor of the object detection device may set the threshold distance to approximately 60 m.

[0109] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is less than or equal to the second reference curvature value (e.g., about 5000 m) and greater than the third reference curvature value (e.g., about 3000 m).

[0110] In an eleventh operation 619, the object recognition device processor may detect whether the curvature is greater than approximately 2000 m. If the curvature is greater than approximately 2000 m, the object recognition device processor may perform a twelfth operation 625. If the curvature is less than or equal to approximately 2000 m, the object recognition device processor may perform a thirteenth operation 623.

[0111] The processor of the object detection device can detect whether the curvature of a lane is less than or equal to the third reference curvature value (e.g., about 3000 m) and greater than a fourth reference curvature value (e.g., about 2000 m).

[0112] In the twelfth operation 625, the processor of the object detection device may set the threshold distance to approximately 50 m.

[0113] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is less than or equal to the third reference curvature value (e.g., about 3000 m) and greater than the fourth reference curvature value (e.g., about 2000 m).

[0114] In the thirteenth operation 623, the object recognition device processor may detect whether the curvature is greater than approximately 1000 m. If the curvature is greater than approximately 1000 m, the object recognition device processor may perform a fourteenth operation 629. If the curvature is less than or equal to approximately 1000 m, the object recognition device processor may perform a fifteenth operation 627.

[0115] The processor of the object detection device can detect whether the curvature of a lane is less than or equal to the fourth reference curvature value (e.g., about 2000 m) and greater than a fifth reference curvature value (e.g., about 1000 m).

[0116] In the fourteenth operation 629, the processor of the object detection device may set the threshold distance to approximately 40 m.

[0117] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is less than or equal to the fourth reference curvature value (e.g., about 2000 m) and greater than the fifth reference curvature value (e.g., about 1000 m).

[0118] In the fifteenth operation 627, the object recognition device processor may detect whether the curvature is greater than approximately 500 m. If the curvature is greater than approximately 500 m, the object recognition device processor may perform the fourth operation 609. If the curvature is less than or equal to approximately 500 m, the object recognition device processor may perform the second operation 605.

[0119] According to one embodiment, the processor of the object detection device can detect whether the curvature (e.g., the radius of curvature) of a lane is less than or equal to the fifth reference curvature value (e.g., about 1000 m) and greater than a sixth reference curvature value (e.g., about 500 m).

[0120] The processor of the object detection device may set the threshold distance to approximately 30 m if the curvature is less than or equal to the fifth reference curvature value (e.g., approximately 1000 m) and greater than the sixth reference curvature value (e.g., approximately 500 m). The processor of the object detection device may set the threshold distance to approximately 20 m if the curvature is less than or equal to the sixth reference curvature value (e.g., approximately 500 m).

[0121] The threshold distance may be greater than the threshold distance if the curvature of a lane on which the host vehicle is located is less than or equal to the fifth reference curvature value (e.g., about 1000 m) and greater than the sixth reference curvature value (e.g., about 500 m).

[0122] Fig. 7 shows an example of lane information where an object is located in an object recognition device or method.

[0123] With reference to Fig. 7, a frame 701 may represent a host vehicle 703 and an object 705. A longitudinal distance position 707 may indicate the position of a point corresponding to the object 705 in a line direction. X may represent an X-axis coordinate value corresponding to the longitudinal distance position. A lateral distance position 709 may indicate the position of the point corresponding to the object 705 in a direction perpendicular to the line direction. Y may represent a Y-axis coordinate value corresponding to the lateral distance position. A point corresponding to the object 705 may include a midpoint of the rearmost line segment in the direction of movement of the object 705 among line segments forming an object box that includes contour points representing the object 705. The midpoint may be referred to as, but is not limited to, a tracking point.A difference value 711 may represent the lateral distance position of a line corresponding to the longitudinal distance position of the object 705.

[0124] The processor of the object detection device may detect the positions of the lines based on at least one of a longitudinal distance position, curvature information, the width of a particular lane or the number of particular lanes, or any combination thereof.

[0125] For example, if the host vehicle 703 is used as the starting point or origin, position information of the lanes may be calculated by the host vehicle 703 as shown in Equation 1. "i" may include an integer between 0 and the number of specific lanes. "w" may denote the width of a specific lane (e.g., about 3.5 m or about 3.7 m). "N" may denote the number of specific lanes. Hereinafter, a value of ,,N2,, be the quotient obtained by dividing "N" by 2, truncating the decimal point. The i-th line may refer to, but is not necessarily limited to, a line detected on the right side of the i-th lane. Position of the i−th line=(W∗(N2−i))+W2

[0126] According to one example, the number of designated lanes may include the number of lanes the host vehicle is located in. For example, if the number of designated lanes is seven, the processor of the object detection device may detect three lanes on each side of the host vehicle. If three lanes are to be detected on each side of the host vehicle and the lane width is set to approximately 3.5 m, the line positions may be as shown in Table 1 below. Table 1 i 0 1 2 3 4 5 6 7 i-te Fahrspur 12.25 m 8.75m 5.25m 1.75m -1.75m -5.25m -8.75m -12.25m

[0127] According to one example, the lateral position of a line corresponding to the longitudinal position 707 in which the object 705 is located may include a value obtained by adding the difference value 711 to the position of the line on the Y-axis (e.g., the position of the (i-th) line calculated in Equation 1). The difference value 711 may be calculated as in Equation 2. The value "c" may denote the curvature value of a lane in which the host vehicle is located. "x" may denote the longitudinal distance position 707 of the object 705. "a" may refer to a smoothing factor. Difference value = c±c2−a∗x2

[0128] According to one example, the processor of the object recognition device may detect the lateral position of the line corresponding to the longitudinal position 707 of the object 705.

[0129] If the lateral position 709 of the object 705 is detected between the lateral position of the nth line corresponding to the longitudinal position 707 of the object 705 and the lateral position of the (n+1)th line corresponding to the longitudinal position 707, the processor of the object recognition device can recognize the object 705 as being located in a lane between the nth line and the (n+1)th line. For example, since the lateral position 709 of the object 705 is detected between the second and third lines in the frame 701, the processor of the object recognition device can recognize that the object 705 is located in the third lane between the second and third lines.

[0130] Through this process, the processor of the object detection device can obtain lane information about a lane in which an object is located.

[0131] Fig. 8 shows an example of the change of a reliability value according to information that changes when a host vehicle changes a lane in an object detection device or method.

[0132] With reference to Fig. 8, a first frame 801, a second frame 811, and a third frame 821 may include objects near a host vehicle 803 detected based on contour points obtained by a LIDAR. The processor of an object detection device may capture frame 801, frame 811, and frame 821 in that order.

[0133] In frame 801, a first object 805 and a second object 807 may be a moving object or an object that may be in a state of motion. In frame 801, a third object 809 may be an object for which it has not yet been determined whether it is a moving object, an object that may be in a state of motion, or an object that may not be in a state of motion.

[0134] If no object other than a moving object or an object capable of being in a state of motion is detected in a lane where an object is located, the confidence value according to information for other objects within the lane may include 0. If an object other than a moving object or an object capable of being in a state of motion is detected in a lane where an object is located, the confidence value according to information for other objects within the lane may include 0.8. If two or more objects other than moving objects or objects capable of being in a state of motion are detected in a lane where an object is located, the confidence value according to information for other objects within the lane may include 1.

[0135] A reliability value according to information for other objects within the lane of the first object 805 of the first frame 801 may be 0, a reliability value according to information for other objects within the lane of the second object 807 of the first frame 801 may be 0.8, and a reliability value according to information for other objects within the lane of the third object 809 of the first frame 801 may be 0.

[0136] In frame 811, the first object 805 may change lanes from a fifth lane to a sixth lane. Despite a lane change of the first object 805, the reliability value may be maintained at 0 according to information for other objects within the lane of the first object 805 in frame 811. Despite a lane change of the first object 805, the reliability value may be maintained at 0.8 according to information for other objects within the lane of the second object 807 in frame 811. When the first object 805 changes lanes, the reliability value may change from 0 to 0.8 according to information for other objects within the lane of the third object 809 in frame 811.

[0137] Even if the first object 805 in frame 811 moves into the lane in which the third object 809 is located, that is, before the first object 805 is detected as a moving object or an object that may be in a moving state, the processor of the object detection device can maintain a confidence value at zero according to information for other objects within the lane of the first object 805 in frame 811. This is because there is no moving object or an object that may be in a moving state in the sixth lane in which the first object 805 is located in frame 811.

[0138] If the third object 809 in the frame 811 is a moving object or an object that may be in a moving state, a reliability value of the first object 805 in the frame 811 may be changed from 0 to 0.8 according to the lane change of the first object 805.

[0139] In frame 821, the second object 807 may change lanes from the seventh lane to the sixth lane. When the second object 807 changes lanes, the reliability value may change from 0 to 0.8 according to information for other objects within the lane of the first object 805 in frame 821. This is because the second object 807 of frame 821, which may be a moving object or an object that may be in a state of movement, is detected within the same lane. Despite a lane change of the second object 807, the reliability value may be maintained at 0.8 according to information for other objects within the lane of the second object 807 in frame 821. When the second object 807 changes lanes, the reliability value may change from 0.8 to 1 according to information for other objects within the lane of the third object 809 in frame 821.This is because the number of moving objects or objects that can be in a state of motion within the sixth lane has increased to two.

[0140] Fig. 9 shows an example of a threshold distance that changes according to the speed of a host vehicle and curvature information in an object detection device or method.

[0141] With reference to Fig. 9, the threshold distance may be determined according to at least one of speed information of the host vehicle or curvature information of a lane in which the host vehicle is located, or any combination thereof. According to one example, an object within the threshold distance may be assigned a confidence value according to information for other objects within the lane of the present disclosure.

[0142] At frame 901, the processor of the object detection device may detect a first threshold distance (e.g., about 20 m) as a first threshold distance 911 of a host vehicle based on a speed of the host vehicle that is less than a reference speed (e.g., about 10 km / h).

[0143] At frame 903, the processor of the object detection device may detect a second threshold distance (e.g., about 30 m) as a second threshold distance 913 of the host vehicle based on the speed of the host vehicle being greater than or equal to the reference speed (e.g., about 10 km / h) and the curvature of a lane in which the host vehicle is located being greater than a first reference curvature (e.g., about 500 m) and less than a second reference curvature (e.g., about 1000 m).

[0144] At frame 905, the processor of the object detection device may detect a third threshold distance (e.g., about 50 m) as a third threshold distance 915 of the host vehicle based on the speed of the host vehicle being greater than or equal to the reference speed (e.g., about 10 km / h) and the curvature of a lane in which the host vehicle is located being greater than a third reference curvature (e.g., about 2000 m), exceeding the second reference curvature (e.g., about 1000 m), and being less than a fourth reference curvature (e.g., about 3000 m).

[0145] At frame 907, the processor of the object detection device may detect a fourth threshold distance (e.g., about 80 m) as a fourth threshold distance 917 of the host vehicle based on the speed of the host vehicle being greater than or equal to the reference speed (e.g., about 10 km / h) and the curvature of a lane in which the host vehicle is located being greater than a fifth reference curvature (e.g., about 10,000 m) that exceeds the fourth reference curvature (e.g., about 3,000 m).

[0146] Fig. 10 shows an example of an object whose reliability is detected within a threshold distance in an object detection apparatus or method.

[0147] With reference to Fig. 10, in frame 1001, a third lane 1003, a fourth lane 1005, and a fifth lane 1007 may represent lanes detected within a threshold distance with respect to the host vehicle.

[0148] When a lane is curved or the host vehicle's lane is changed, as in the lane shown in frame 1001, lane information may be inaccurate. Errors in the lane information may result in errors in reliability according to information for other objects within the lane. According to one example, to reduce errors in object classification due to errors in reliability, the processor of the object detection device may assign reliability according to information for other objects within the lane only to objects included within a threshold distance. Object classification may include detecting whether an object is a moving object, an object that may be in a state of motion, or an object that may not be in a state of motion.

[0149] According to one example, the lane information of an object within the threshold distance may have an accuracy of a reference value or more even when a lane is curved or the host vehicle's lane is changed, as in the lane shown in frame 1001.

[0150] Fig. 11 shows an example of a change in a reliability value according to information that changes depending on the positions of other objects in an object recognition device or an object recognition method.

[0151] With reference to Fig. 11, the processor of the object recognition device may capture frames in the order of a first frame 1101 and a second frame 1111. The first frame 1101 may include a host vehicle 1103, a first object 1105, a second object 1107, and a third object 1109. The second frame 1111 may include the host vehicle 1103, the first object 1105, and the second object 1107.

[0152] The first object 1105 of the first frame 1101 may be... before the first object 1105 is recognized as a moving object or as an object that may be in a state of motion. The second object 1107 of the first frame 1101 and the third object 1109 of the first frame 1101 may be recognized as a moving object or as an object that may be in a state of motion.

[0153] According to one example, in the first frame 1101, the reliability value of the first object 1105 may be recognized as 1 according to the information for other objects within the lane because other objects (e.g., the second object 1107 of the first frame 1101 and the third object 1109 of the first frame 1101) are recognized as a moving object or an object that may be in a moving state within the same lane.

[0154] According to one example, in the second frame 1111, the third object 1109 detected in the first frame 1101 cannot be detected in the lane where the first object 1105 of the second frame 1111 is located when the host vehicle is moving. The reliability value according to the information about other objects in a lane where the first object 1105 of the second frame 1111 is located can be recognized as 0.8 because another object (e.g., the second object 1107 of the second frame 1111) is detected as a moving object or an object that may be in a moving state within the same lane.

[0155] The reliability value according to information about other objects in the lane may change when the host vehicle moves or the lane is changed.

[0156] Fig. 12 shows a computer system relating to an object recognition device or method.

[0157] According to one embodiment of the present disclosure, an object detection device includes a LIDAR and a processor.

[0158] According to one example, the processor may detect longitudinal position information representing a distance in a line direction from a host vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction from the host vehicle to the object through the LIDAR; detect lane information representing a lane in which the object is located according to at least one of the lateral position information or positions of lines located on both sides of the host vehicle, or any combination thereof, based on the longitudinal position information falling within a range according to a threshold distance; detect whether at least one other object, different from the object and which may be considered a moving object or an object that may be in a state of movement, is located within the lane in which the object is located;assign a first value as a reliability value of the object if the at least one other object is located within the lane in which the object is located, assign a second value, which is smaller than the first value, as the reliability value if the at least one other object is not located within the lane in which the object is located, and recognize the object as a moving object or as an object that may be in a state of movement based on the reliability value.

[0159] According to one example, the processor may detect curvature information indicating a curvature of a lane in which the host vehicle is located. The positions of the lines may be detected based on at least one of the longitudinal position information, the curvature information, a width of a particular lane, or a number of the particular lanes, or any combination thereof.

[0160] According to one example, based on the lateral position information, the processor may detect that the object is located between a position of a first line and a position of a second line adjacent to the first line. The lane information may include a lane located between the first line and the second line.

[0161] According to one example, a threshold distance of longitudinal position information detected by the host vehicle traveling at a first speed may be less than a threshold distance of longitudinal position information detected by the host vehicle traveling at a second speed that is higher than the first speed.

[0162] According to one example, a threshold distance of longitudinal position information detected by the host vehicle traveling in a lane having a first curvature may be less than a threshold distance of longitudinal position information detected by the host vehicle traveling in a lane having a second curvature that is greater than a first curvature.

[0163] According to one example, the processor may detect curvature information indicative of a curvature of the lane in which the host vehicle is located and detect the longitudinal position information and the lateral position information based on determining that the curvature information is greater than a certain curvature value.

[0164] According to one example, the processor may assign a third value as a confidence value if a first number of other objects that are detected as moving objects or objects that may be in a state of movement are within the lane in which the object is located, and assign a fourth value that is greater than the third value as a confidence value if a second number of other objects that is greater than the first number are detected within the lane in which the object is located.

[0165] According to one example, the processor may change the reliability value from the second value to the first value when the lane information of the object changes because the object moves from a first lane in which the other object is not located to a second lane in which the other object is located.

[0166] According to one example, the processor may maintain the confidence value even if at least one other object moves into the lane in which the object is located before the object is detected as a moving object or before the object is detected as an object that may be in a state of motion.

[0167] According to one example, the processor may assign to the object an identifier indicating that the object is a moving object or an object that may be in a state of motion based on determining that a score value representing a probability that the object detected based on the confidence value is a moving object or an object that may be in a state of motion is greater than a score value representing a probability that the object is an object that may not be in a state of motion.

[0168] According to one embodiment of the present disclosure, an object detection method includes detecting, by the LIDAR, longitudinal position information representing a distance in a line direction from a host vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction from the host vehicle to the object; detecting lane information representing a lane in which the object is located according to at least one of the lateral position information or positions of lines located on both sides of the host vehicle, or any combination thereof based on determining that the longitudinal position information falls within a range according to a threshold distance; detecting whether at least one other object different from the object and identified as a moving object or an object,which may be in a state of motion, is detected, is located within the lane in which the object is located, assigning a first value as a reliability value of the object if the at least one other object is located within the lane in which the object is located, assigning a second value, which is smaller than the first value, as the reliability value if the at least one other object is not within the lane in which the object is located, and detecting the object as a moving object or as an object that may be in a state of motion, based on the reliability value.

[0169] According to one example, the object detection method may further include detecting curvature information indicating a curvature of the lane in which the host vehicle is located. The positions of the lines may be detected based on at least one of the longitudinal position information, the curvature information, a width of a particular lane, or a number of the particular lanes, or any combination thereof.

[0170] According to one example, detecting whether at least one other object, different from the object and identified as a moving object or an object capable of being in a state of movement, is located within the lane in which the object is located may include detecting, based on the lateral position information, that the object is located between a position of a first line and a position of a second line adjacent to the first line. The lane information may include a lane located between the first line and the second line.

[0171] According to one example, a threshold distance of longitudinal position information detected by the host vehicle traveling at a first speed may be less than a threshold distance of longitudinal position information detected by the host vehicle traveling at a second speed that is higher than the first speed.

[0172] According to one example, a threshold distance of longitudinal position information detected by the host vehicle traveling in a lane having a first curvature may be less than a threshold distance of longitudinal position information detected by the host vehicle traveling in a lane having a second curvature that is greater than a first curvature.

[0173] According to one example, the object detection method may further include detecting curvature information indicating a curvature of the lane in which the host vehicle is located, and detecting the longitudinal position information and the lateral position information based on determining that the curvature information is greater than a certain curvature value.

[0174] According to one example, the object detection method may further comprise assigning a third value as the reliability value when a first number of other objects that are detected as a moving object or objects that may be in a state of movement are detected within the lane in which the object is located, and assigning a fourth value that is greater than the third value as the reliability value when a second number of other objects that is greater than the first number is detected within the lane in which the object is located.

[0175] According to one example, the object detection method may further comprise changing the reliability value from the second value to the first value when the lane information of the object changes because the object moves from a first lane in which the other object is not located to a second lane in which the other object is located.

[0176] According to one example, the object detection method may further comprise maintaining the confidence value even if at least one other object moves into the lane in which the object is located before the object is detected as a moving object or before the object is detected as an object that may be in a state of motion.

[0177] According to one example, the object detection method may further comprise assigning to the object an identifier indicating that the object is a moving object or an object that may be in a state of motion based on determining that a score value representing a probability that the object detected based on the reliability value is a moving object or an object that may be in a state of motion is greater than a score value representing a probability that the object is an object that may not be in a state of motion.

[0178] With reference to Fig. 12, a computer system 1200 may include at least a processor 1210, a memory 1230, a user interface input device 1240, a user interface output device 1250, a storage 1260, and a network interface 1270 interconnected via a bus 1220.

[0179] Processor 1210 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in memory 1230 and / or storage 1260. Memory 1230 and storage 1260 may include various types of volatile or non-volatile storage media. For example, memory 1230 may include ROM (Read Only Memory) 1231 and RAM (Random Access Memory) 1232.

[0180] Thus, the operations of the method or algorithm described in connection with the features disclosed herein may be embodied directly in hardware or in a software module executed by processor 1210, or in a combination thereof. The software module may be located on a storage medium (i.e., memory 1230 and / or storage 1260) such as RAM, flash memory, ROM, EPROM, EEPROM, register, hard disk, removable disk, and CD-ROM.

[0181] The exemplary storage medium may be coupled to the processor 1210, and the processor 1210 may read information from the storage medium and store information on / in the storage medium. Alternatively, the storage medium may be integrated with the processor 1210. The processor and the storage medium may be housed in an application-specific integrated circuit (ASIC). The ASIC may be housed in a user terminal. Alternatively, the processor and the storage medium may be housed as separate components in the user terminal.

[0182] The above description is merely illustrative of the technical idea of ​​the present disclosure, and various modifications and variations can be made by a person skilled in the art to which the present disclosure belongs without departing from the essential characteristics of the present disclosure.

[0183] Accordingly, the embodiment disclosed in the present disclosure is not intended to limit the technical idea of ​​the present disclosure, but to describe the present disclosure, and the scope of the technical idea of ​​the present disclosure is not limited by the embodiment. The scope of the present disclosure should be interpreted by the following claims, and all technical ideas within the equivalent scope should be construed as being included in the scope of the present disclosure.

[0184] The present technology can increase the accuracy of a determination of whether an object is a moving object or an object that may be in a state of motion by using at least one other object in the same lane as the object.

[0185] Furthermore, the present technology can detect whether an object having a portion occluded is a moving object or an object that may be in a state of motion using at least one other object in the same lane as the object.

[0186] Furthermore, the present technology can detect whether an object in a stationary state is a moving object or an object that may be in a state of motion using at least one other object in the same lane as the object.

[0187] Furthermore, the present technology can enhance the user experience by improving the accuracy of determining whether an object is a moving object or an object that may be in a state of motion.

[0188] Furthermore, the present technology can improve the performance of autonomous driving by increasing the accuracy of determining whether an object is a moving object or an object that may be in a state of motion.

[0189] In addition, various effects may be provided which are understood directly or indirectly through the disclosure.

[0190] Although the present disclosure has been described above with reference to various examples and the accompanying drawings, the present disclosure is not limited thereto, but can be variously modified and altered by a person skilled in the art to which the present disclosure belongs, without departing from the spirit and scope of the present disclosure as claimed in the following claims.

[0191] Fig. 3. 301 frames 303 Host vehicle 305 first area 307 second area 309 third area

[0192] Fig. 4. 401 screen 403 first object 411 frames 413 Host vehicle 415 first object 417 second object

[0193] Fig. 7. 703 Host vehicle 705 Object 707 Longitudinal distance position 709 Page spacing position 711 Difference value

[0194] Fig. 8. 801 first frame 803 Host Vehicle 805 first object 807 second object 809 third object 811 Frame 821 third frame

[0195] Fig. 9. 901 Frame 903 Frame 905 Frame 907 Frame 911 first threshold distance 913 second threshold distance 915 third threshold distance 917 fourth threshold distance

[0196] Fig. 10. 1001 Frame 1003 third lane 1005 fourth lane 1007 fifth lane

[0197] Fig. 11. 1101 first frame 1103 Host Vehicle 1105 first object 1107 second object 1109 third object 1111 second frame

[0198] Fig. 12. 1200 computer system

Claims

[1] Device comprising: a sensor; and a processor, wherein the processor is configured to: to determine, based on sensing information from the sensor, longitudinal position information representing a distance in a line direction from a vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction; based on a determination that the longitudinal position information falls within a range according to a threshold distance, determine lane information representing a lane in which the object is located, wherein the lane information is determined based on at least one of: the lateral position information or positions of lines located on both sides of the vehicle; to determine whether at least one other object, different from the object, which is detected as a moving object or as a stationary object, is located within the lane in which the object is located; assign a reliability value to the object, wherein a first value corresponds to the reliability value for the object based on the at least one other object being within the lane in which the object is located, or a second value, which is less than the first value, corresponds to the reliability value for the object based on the at least one other object not being within the lane in which the object is located; to determine the object as a moving object or as a stationary object based on the reliability value; and output a signal indicating that the object is a moving object or a signal indicating that the object is a stationary object. [2] The apparatus of claim 1, wherein the processor is configured to: detect curvature information indicating a curvature of a lane in which the vehicle is located, and to detect the positions of the lines based on at least one of: the longitudinal position information, the curvature information, a width of a particular lane, or a number of the particular lanes. [3] The apparatus of claim 1, wherein the processor is configured to detect, based on the lateral position information, that the object is located between a position of a first line and a position of a second line adjacent to the first line, and wherein the lane information is configured to indicate a lane located between the first line and the second line. [4] The apparatus of claim 1, wherein a threshold distance of longitudinal position information detected by the vehicle traveling at a first speed is arranged to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling at a second speed higher than the first speed. [5] The apparatus of claim 1, wherein a threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a first curvature is configured to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a second curvature that is greater than a first curvature. [6] The apparatus of claim 1, wherein the processor is configured to: to determine curvature information indicating a curvature of a lane in which the vehicle is located; and to determine the longitudinal position information and the lateral position information based on the curvature information being greater than a certain curvature value. [7] Device according to claim 1, wherein: a third value corresponds to the confidence value for the object when a first number of other objects that are recognized as moving objects or objects that may be in a state of movement are detected within the lane in which the object is located; and a fourth value, which is greater than the third value, corresponds to the confidence value for the object when a second number of other objects, which is greater than the first number, is detected within the lane in which the object is located. [8] The apparatus of claim 1, wherein the processor is configured to change the reliability value from the second value to the first value based on the lane information of the object that is changed because the object moves from a first lane in which no other object is located to a second lane in which the other object is located. [9] The apparatus of claim 1, wherein the processor is configured to maintain the reliability value even if at least one other object moves into the lane in which the object is located before the object is detected as a moving object or before the object is detected as a stationary object. [10] The apparatus of claim 1, wherein the processor is configured to assign to the object an identifier indicating that the object is a moving object or a stationary object that may be in a state of motion based on a determination that a score value representing a probability that the object detected based on the reliability value is a moving object or a stationary object that may be in a state of motion is greater than a score value representing a probability that the object is a stationary object that may not be in a state of motion. [11] Method comprising: Determining, by a processor and based on sensing information from a sensor, longitudinal position information representing a distance in a line direction from a vehicle to an object and lateral position information representing a distance in a direction perpendicular to the line direction; Determining, based on a determination that the longitudinal position information falls within a range according to a threshold distance, lane information representing a lane in which the object is located, wherein the lane information is determined based on at least one of: the lateral position information or positions of lines located on both sides of the vehicle; Determining whether at least one other object that is different from the object and is detected as a moving or stationary object is located within the lane in which the object is located; Assigning a reliability value to the object, wherein a first value corresponds to the reliability value for the object based on the at least one other object being within the lane in which the object is located, or a second value, which is less than the first value, corresponds to the reliability value for the object based on the at least one other object not being within the lane in which the object is located; Determining, based on the reliability value, the object as a moving object or as a stationary object; and Output a signal indicating that the object is a moving object or a signal indicating that the object is a stationary object. [12] The method of claim 11, further comprising: Detecting curvature information indicating a curvature of the lane in which the vehicle is located; and Detecting the positions of the lines based on at least one of: the longitudinal position information, the curvature information, a width of a particular lane, or a number of the particular lanes. [13] The method of claim 11, wherein determining whether at least one other object is within the lane in which the object is located comprises: Determining, based on the transverse position information, that the object is located between a position of a first line and a position of a second line that is adjacent to the first line, wherein the lane information is configured to indicate a lane located between the first line and the second line. [14] The method of claim 11, wherein a threshold distance of longitudinal position information detected by the vehicle traveling at a first speed is arranged to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling at a second speed higher than the first speed. [15] The method of claim 11, wherein a threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a first curvature is configured to be smaller than a threshold distance of longitudinal position information detected by the vehicle traveling in a lane having a second curvature that is greater than a first curvature. [16] The method of claim 11, further comprising: Determining curvature information indicating a curvature of a lane in which the vehicle is located; and Determining the longitudinal position information and the lateral position information based on the curvature information being greater than a certain curvature value. [17] The method of claim 11, wherein: a third value corresponds to the confidence value for the object when a first number of other objects that are recognized as moving objects or objects that may be in a state of movement are detected within the lane in which the object is located; and a fourth value, which is greater than the third value, corresponds to the confidence value for the object when a second number of other objects, which is greater than the first number, is detected within the lane in which the object is located. [18] The method of claim 11, further comprising: Changing the confidence value from the second value to the first value based on the lane information of the object changing because the object moves from a first lane in which another object is not located to a second lane in which the other object is located. [19] The method of claim 11, further comprising: Maintain the confidence value even if at least one other object moves into the lane in which the object is located before the object is detected as a moving object or before the object is detected as a stationary object. [20] The method of claim 11, further comprising: Assigning to the object an identifier indicating that the object is a moving object or a stationary object that may be in a state of motion based on a determination that a score value representing a probability that the object detected based on the confidence value is a moving object or a stationary object that may be in a state of motion is greater than a score value representing a probability that the object is a stationary object that may not be in a state of motion.

Images