Short-distance cutting-in target recognition device and recognition method

Abstract

The present invention discloses a short-distance cut-in target recognition device and a recognition method thereof. The short-distance cut-in target recognition device includes: an Occupancy Distance Map (ODM) information calculator configured to calculate ODM information based on information of the host vehicle and surrounding objects; a trajectory information calculator configured to calculate trajectory information based on information of the host vehicle and surrounding objects; and a short-distance cut-in target selector configured to select a short-distance cut-in target based on the ODM information and the trajectory information.

Description

[0001] Cross - reference to related applications

[0002] This application claims priority to Korean Patent Application No. 10 - 2019 - 0143943, filed on November 12, 2019, which is hereby incorporated by reference in its entirety. Technical Field

[0003] The present disclosure relates to a short - distance cut - in target recognition device and a recognition method. Background Art

[0004] Generally, vehicles as transportation devices have evolved into intelligent vehicles that can utilize recently developed information and communication technologies to provide improved safety and convenience, and to enhance fuel efficiency and performance.

[0005] However, since intelligent vehicles include additional functions such as entertainment systems, air filters, and convenience devices, in addition to operating the control devices for driving, the driver also needs to operate other additional control devices, thus increasing the risk of accidents due to driver carelessness.

[0006] Therefore, recently, various studies have been conducted on safety devices that can prevent or avoid vehicle collisions.

[0007] Vehicle collision avoidance devices include an adaptive cruise control system, a forward vehicle collision warning system, a lane departure warning system, etc., and these vehicle collision avoidance devices are mainly used in high - speed driving situations to prevent serious accidents. Most of the technologies used in such vehicle collision avoidance devices involve detecting obstacles at a long distance in high - speed driving situations.

[0008] However, for most actual traffic accidents, due to urban traffic congestion, more than 70% of the accidents mainly occur at low speeds equal to or less than about 30 km / h. Therefore, traditional collision avoidance devices are not suitable for accurately identifying an oncoming vehicle that cuts in at a low speed within a short distance and preventing collisions.

[0009] For example, when an oncoming vehicle cuts in at a low speed within a short distance, the host vehicle may not accurately identify the corner radar information due to the noise included in the corner radar information, or may not accurately identify the oncoming vehicle that cuts in at a low speed within a short distance due to inertial coasting, thus misjudging the situation and hitting the oncoming vehicle.

[0010] The problem with the current Smart Cruise Control (SCC) system is that the system cannot identify a forward vehicle that deviates from the recognition area of the front radar and the front camera.

[0011] On congested roads, other vehicles often cut in over short distances, and there is also a problem that when the vehicle ahead deviates from the recognition area under corresponding circumstances, the vehicle ahead cannot decelerate accordingly, resulting in a risk of collision.

[0012] Therefore, in the future, there is a need for a device that can identify short-distance cut-in targets by accurately judging various situations of other vehicles cutting in at low speeds over short distances, so as to improve reliability and safety. Summary of the Invention

[0013] The present disclosure relates to a short-distance cut-in target recognition device. Specific embodiments relate to a short-distance cut-in target recognition device and a recognition method thereof for determining a short-distance cut-in target by using occupancy distance map (ODM) information and track information.

[0014] Embodiments of the present disclosure provide a short-distance cut-in target recognition device and a recognition method thereof for matching ODM information corrected based on the driving state of the host vehicle with a plurality of candidate tracks selected based on track information to identify a short-distance cut-in target vehicle, thereby improving reliability and safety.

[0015] The technical problems solved by the embodiments are not limited to the above technical problems, and according to the following description, other technical problems not described herein will become apparent to those skilled in the art.

[0016] According to an embodiment of the present disclosure, a short-distance cut-in target recognition device includes: an occupancy distance map (ODM) information calculator configured to calculate ODM information based on the host vehicle and surrounding object information; a track information calculator configured to calculate track information based on the host vehicle and surrounding object information; a short-distance cut-in target selector configured to select a short-distance cut-in target based on the ODM information and the track information. The short-distance cut-in target selector determines whether the host vehicle satisfies the short-distance cut-in target recognition entry condition, validates the ODM information and removes invalid ODM information when the host vehicle satisfies the short-distance cut-in target recognition entry condition, corrects the ODM information based on the driving state of the host vehicle, selects an ODM track candidate group including a plurality of candidate tracks based on the track information, matches each candidate track and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify short-distance cut-ins of the candidate tracks, determines a short-distance cut-in target from the short-distance cut-in candidate tracks, and outputs the determined short-distance cut-in target information.

[0017] According to another embodiment of the present disclosure, a method for a short-distance cut-in target recognition device to recognize a short-distance cut-in target is provided. The short-distance cut-in target recognition device includes a short-distance cut-in target selector that selects a short-distance cut-in target based on ODM information and track information. The method includes: the short-distance cut-in target selector determines whether the vehicle itself meets the short-distance cut-in target recognition entry condition; when the vehicle itself meets the short-distance cut-in target recognition entry condition, the short-distance cut-in target selector verifies the validity of the ODM information and removes the invalid ODM information; the short-distance cut-in target selector corrects the ODM information based on the driving state of the vehicle itself; the short-distance cut-in target selector selects an ODM track candidate group including a plurality of candidate tracks based on the track information; the short-distance cut-in target selector matches each candidate track and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify the short-distance cut-in of the candidate track; the short-distance cut-in target selector determines the short-distance cut-in target from the short-distance cut-in candidate tracks; and the short-distance cut-in target selector outputs the determined short-distance cut-in target information.

[0018] According to an embodiment of the present disclosure, a program for executing this method is recorded in a computer-readable recording medium.

[0019] According to an embodiment of the present disclosure, a vehicle includes: a sensing device configured to sense the vehicle itself and surrounding objects; and a short-distance cut-in target recognition device configured to recognize a short-distance cut-in vehicle based on the information of the vehicle itself and surrounding objects received from the sensing device. The short-distance cut-in target recognition device includes: an occupancy distance map (ODM) information calculator configured to calculate ODM information based on the information of the vehicle itself and surrounding objects; a track information calculator configured to calculate track information based on the information of the vehicle itself and surrounding objects; and a short-distance cut-in target selector configured to determine whether the vehicle itself meets the short-distance cut-in target recognition entry condition, verify the validity of the ODM information and remove the invalid ODM information when the vehicle itself meets the short-distance cut-in target recognition entry condition, correct the ODM information based on the driving state of the vehicle itself, select an ODM track candidate group including a plurality of candidate tracks based on the track information, match each candidate track and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify the short-distance cut-in of the candidate track, determine the short-distance cut-in target from the short-distance cut-in candidate tracks, and output the determined short-distance cut-in target information. Description of the Drawings

[0020] The drawings included to provide a further understanding of the present disclosure and incorporated into this application and constituting a part of this application illustrate embodiments of the present disclosure and are used together with the description to explain the principles of the present disclosure. In the drawings:

[0021] Figure 1is a block diagram for explaining a short-distance cut-in target recognition device according to the present disclosure;

[0022] Figure 2 is a diagram for explaining the process of recognizing the entry conditions for a short-distance cut-in target recognition;

[0023] Figure 3 and Figure 4 is a diagram for explaining the process of verifying the validity of occupancy distance map (ODM) information;

[0024] Figure 5 and Figure 6 is a diagram for explaining the process of correcting ODM information;

[0025] Figure 7 is a diagram for explaining the process of selecting a candidate group of ODM tracks;

[0026] Figure 8 and Figure 9 is a view for explaining the process of matching ODM information and candidate tracks with each other;

[0027] Figure 10 is a diagram for explaining the process of recognizing a short-distance cut-in of a candidate track;

[0028] Figure 11 is a diagram for explaining the process of determining a short-distance cut-in target;

[0029] Figures 12 to 14 is a diagram for explaining the application of a short-distance cut-in target recognition device according to an embodiment of the present disclosure; and

[0030] Figure 15 is a flowchart for explaining the method of recognizing a short-distance cut-in target according to the present disclosure. Detailed Description of the Specific Embodiment

[0031] The exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present disclosure. However, the present disclosure can be implemented in various different forms and is not limited to these embodiments. To clearly describe the present disclosure, parts irrelevant to the description are omitted in the drawings, and the same reference numerals in the specification denote the same elements.

[0032] Throughout the specification, unless otherwise clearly defined, those of ordinary skill in the art will understand that the terms "include", "comprise", and "have" are defaultly interpreted as inclusive or open-ended, rather than exclusive or closed. In addition, terms such as "unit" and "module" disclosed in the specification refer to a unit for processing at least one function or operation, which can be implemented by hardware, software, or a combination of hardware and software.

[0033] Throughout the specification, when a component "includes" a certain component, unless there is a different disclosure, this means that the component may further include another component without excluding the other component. The same reference numerals will be used throughout the drawings to denote the same components.

[0034] Hereinafter, a short-distance cut-in target recognition device and a recognition method applicable to the present disclosure will be described in detail with reference to Figures 1 to 15 A short-distance cut-in target recognition device and a recognition method applicable to the present disclosure will be described in detail.

[0035] Figure 1 is a block diagram for explaining a short-distance cut-in target recognition device according to the present disclosure.

[0036] As Figure 1 shown, the short-distance cut-in target recognition device according to the present disclosure may include an Occupancy Distance Map (ODM) information calculator 100, a trajectory information calculator 200, and a short-distance cut-in target selector 300.

[0037] Here, the ODM information calculator 100 may calculate ODM information based on information about the host vehicle and surrounding objects.

[0038] That is, the ODM information calculator 100 may be an element for calculating ODM information, and may calculate information (position, relative speed, etc.) of an object existing in a lattice-type space as a concept of free space.

[0039] The ODM information may include a grid map corresponding to the ODM object and a surrounding object detection area including a plurality of detection points.

[0040] Here, the detection points may use only valid detection information from the detection information transmitted from the radar.

[0041] For example, the detection information for determining the ODM may include distance information, speed information, angle information, and signal strength information.

[0042] The ODM object may be a cluster, and may be a detection group determined to be an object or a vehicle.

[0043] The ODM object may use the detection information output to the ODM area to track a moving or stationary object.

[0044] Then, the grid map may divide the front longitudinal and lateral regions into six parts, but is not limited thereto.

[0045] For example, the grid map may include two inner regions IR, two first outer regions ER1, and two second outer regions ER2.

[0046] The width of an internal area can be calculated using IR = vehicle width / 2 + α (where IR (internal area) is the width of the internal area and α is a margin value considering side mirrors).

[0047] Here, the vehicle width can vary according to the specifications of the present vehicle, and α can be from approximately 0.5 m to approximately 1.5 m, but the present invention is not limited thereto.

[0048] The width of a first external area can be calculated using ER1 = lane width / 2 - IR (where ER1 (first external area) is the width of the first external area and IR is the width of the internal area).

[0049] Here, the lane width can be from approximately 3 m to approximately 4 m, but the present invention is not limited thereto.

[0050] The width of a second external area can be from approximately 0.2 m to approximately 0.8 m.

[0051] The length of the internal area can be the same as the length of each of the first and second external areas.

[0052] For example, the length of each of the internal area, the first external area, and the second external area can be from approximately 15 m to approximately 25 m in the direction forward from the front surface of the present vehicle.

[0053] Then, the trajectory information calculator 200 can calculate trajectory information based on the present vehicle and surrounding object information.

[0054] Here, the trajectory information calculator 200 can calculate information typically output from front recognition sensors such as radar, LIDAR, or camera sensors, as well as information (position, relative speed, etc.) of objects recognized as having an independent form (vehicle / pedestrian / building, etc.).

[0055] Then, the short - distance cut - in target selector 300 can identify short - distance cut - in vehicles based on ODM information and trajectory information.

[0056] That is, the short-distance cut-in target selector 300 can determine whether the vehicle meets the short-distance cut-in target recognition entry condition. When the vehicle meets the short-distance cut-in target recognition entry condition, it can verify the validity of the ODM information and remove the invalid ODM information, can correct the ODM information based on the driving state of the vehicle, can select an ODM track candidate group including multiple candidate tracks based on the track information, can match each candidate track and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify the short-distance cut-in of the candidate track, can determine the short-distance cut-in target from the short-distance cut-in candidate tracks, and can output the determined short-distance cut-in target information.

[0057] Here, when determining whether the vehicle meets the short-distance cut-in target recognition entry condition, the short-distance cut-in target selector 300 can receive the driving state information of the vehicle including the offset information, heading information, and slalom information of the vehicle, and can determine whether the driving state of the vehicle meets the short-distance cut-in target recognition entry condition based on the driving state information of the vehicle.

[0058] When verifying the validity of the ODM information, the short-distance cut-in target selector 300 can determine the attribute of the ODM information to identify the ODM information from the fixed obstacle, and can determine that the identified ODM information is invalid.

[0059] When verifying the validity of the ODM information, the short-distance cut-in target selector 300 can identify the ODM information measured by the target of the vehicle lane in front from the ODM information, and can determine that the identified ODM information is invalid.

[0060] Then, when correcting the ODM information, the short-distance cut-in target selector 300 can correct the ODM information based on at least one of the offset information and heading information of the vehicle.

[0061] When correcting the ODM information, the short-distance cut-in target selector 300 can rearrange the ODM information including the grid map based on the coordinate system of the driving lane.

[0062] Then, when selecting the ODM track candidate group, the short-distance cut-in target selector 300 can select the tracks in the left lane and the right lane based on the vehicle lane from the multiple sensor fusion tracks as the ODM track candidate group based on the track information.

[0063] Then, when determining the short-distance cut-in of the candidate track, the short-distance cut-in target selector 300 can add a margin value to the length and width of each candidate track to set the candidate track matching area, and can identify the short-distance cut-in of the candidate track based on the set candidate track matching area.

[0064] Here, when setting the candidate track matching area, the short-distance cut-in target selector 300 can add the length of the candidate track and the length margin to set the longitudinal area of the candidate track matching area, and can add the width of the candidate track and the width margin to set the lateral area of the candidate track matching area.

[0065] When determining the short-distance cut-in of the candidate track, the short-distance cut-in target selector 300 can add the course width margin value to the width of each candidate track to set the candidate track matching area, and can identify the short-distance cut-in of the candidate track based on the set candidate track matching area.

[0066] Here, when setting the candidate track matching area, the short-distance cut-in target selector 300 can multiply the length of the candidate track and the course angle of the candidate track to calculate the course width margin value, and can add the calculated course width margin value and the width of the candidate track to set the lateral area of the candidate track matching area.

[0067] Then, when determining the short-distance cut-in of the candidate track, the short-distance cut-in target selector 300 can match each candidate track and the ODM object corresponding to each candidate track with the grid map, and can judge the short-distance cut-in state of each candidate track based on the matching.

[0068] Here, when judging the short-distance cut-in state of each candidate track, the short-distance cut-in target selector 300 can judge the short-distance cut-in state of the candidate track as one of the Not Detected state, New state, Update state, Track Coasting state, ODM Coasting state, and Time Coasting state based on the matching.

[0069] Then, when determining the short-distance cut-in target, the short-distance cut-in target selector 300 can determine the candidate track closest to the host vehicle among the candidate tracks identified as short-distance cut-in tracks in the left lane and the right lane based on the host vehicle lane as the short-distance cut-in target.

[0070] For example, when determining the short-distance cut-in target, the short-distance cut-in target selector 300 can select the first candidate track closest to the host vehicle among the candidate tracks identified as short-distance cut-in tracks in the left lane based on the host vehicle lane, can select the second candidate track closest to the host vehicle among the candidate tracks identified as short-distance cut-in tracks in the right lane based on the host vehicle lane, and can determine the two candidate tracks including the first candidate track and the second candidate track as the short-distance cut-in target.

[0071] When outputting the determined short-distance cut-in target information, the short-distance cut-in target selector 300 may calculate position information and relative speed information based on the measurement points of the short-distance cut-in target, and may output short-distance cut-in target information including the calculated position information and relative speed information.

[0072] As described above, according to the present disclosure, the ODM information corrected based on the driving state of the host vehicle and the multiple candidate tracks selected based on the track information may be matched with each other to identify a short-distance cut-in target, thereby improving reliability and safety.

[0073] Therefore, according to the present disclosure, various situations of an oncoming vehicle cutting in at a low speed within a short distance can be accurately determined, thereby improving reliability and safety.

[0074] Figure 2 It is a diagram for explaining the process of identifying the entry conditions for short-distance cut-in target recognition.

[0075] As Figure 2 shown, first, the short-distance cut-in target selector according to the present disclosure may determine whether the host vehicle satisfies the short-distance cut-in target recognition entry conditions to identify a short-distance cut-in vehicle.

[0076] That is, when the driving state of the host vehicle satisfies the short-distance cut-in target recognition entry conditions, the short-distance cut-in target selector may execute the process of identifying a short-distance cut-in vehicle.

[0077] Therefore, when determining whether the host vehicle satisfies the short-distance cut-in target recognition entry conditions, the short-distance cut-in target selector may receive the driving state information of the host vehicle including the offset information, heading information, and turning information of the host vehicle, and may determine whether the driving state of the host vehicle satisfies the short-distance cut-in target recognition entry conditions based on the driving state information of the host vehicle.

[0078] Here, when determining whether the host vehicle satisfies the short-distance cut-in target recognition entry conditions, the short-distance cut-in target selector may determine whether the driving state of the host vehicle satisfies at least one of the offset driving condition, heading driving condition, and turning driving condition.

[0079] For example, when determining whether the offset driving condition is satisfied, the short-distance cut-in target selector may determine whether the offset information of the host vehicle satisfies the offset driving condition with an effective condition of about -35 cm to about 35 cm and a non-determined condition of about -40 cm to about 40 cm.

[0080] In another example, when determining whether the heading driving condition is satisfied, the short-distance cut-in target selector can determine whether the heading information of the vehicle satisfies the valid condition of approximately -1.8 degrees to approximately 1.8 degrees and does not satisfy the heading driving condition with the condition of approximately -2.0 degrees to approximately 2.0 degrees.

[0081] In yet another example, when determining whether the turning driving condition is satisfied, the short-distance cut-in target selector can determine whether the turning information of the vehicle satisfies the valid turning information of approximately -1.0 degree / second to approximately 1.0 degree / second and does not satisfy the turning driving condition with the condition of approximately -1.2 degrees / second to approximately 1.2 degrees / second.

[0082] When receiving the offset information, heading information, and turning information of the vehicle, the short-distance cut-in target selector can determine whether the lane information of the vehicle is valid when receiving the lane information of the vehicle, and can receive the offset information, heading information, and turning information of the vehicle when the lane information of the vehicle is valid.

[0083] The short-distance cut-in target selector can identify a short-distance cut-in target based on the driving speed of the vehicle, the relative speed of the target, and whether the target is moving.

[0084] For example, when the driving speed of the vehicle is equal to or less than approximately 20 kph (approximately 12.5 mph), the short-distance cut-in target selector can identify a short-distance cut-in target.

[0085] In another example, when the relative speed of the target is from approximately -10 m / s to approximately 10 m / s, the short-distance cut-in target selector can identify a short-distance cut-in target.

[0086] In yet another example, when the target is a moving target that moves within approximately 20 m or a stationary target that has previously moved, the short-distance cut-in target selector can identify a short-distance cut-in target.

[0087] In this way, according to the present disclosure, it is possible to determine the short-distance cut-in target recognition entry condition to eliminate the possibility of misjudgment when using ODM information to determine the cut-in.

[0088] According to the present disclosure, when the vehicle is driving while leaning to one side in the driving lane, when the vehicle is driving while tilted in the driving lane, or when the vehicle is turning in the driving lane, if the short-distance cut-in target recognition entry condition is satisfied, the process of recognizing a short-distance cut-in target can be executed.

[0089] For example, as Figure 2 shown, the basic conditions for short-distance cut-in target recognition entry are designed as follows, but are not limited thereto.

[0090] Here, according to the present disclosure, when the lane information is valid, the offset information and the heading information of the vehicle with respect to the driving lane can be received.

[0091] First, the driving condition of the vehicle offset can have a valid condition of about -35 cm to about 35 cm and an undetermined condition of about -40 cm to about 40 cm.

[0092] Second, the driving condition of the vehicle heading can have a valid condition of about -1.8 degrees to about 1.8 degrees and an undetermined condition of about -2.0 degrees to about 2.0 degrees.

[0093] Third, the driving condition of the vehicle turning can have a valid condition of about -1.0 degree / second to about 1.0 degree / second and an undetermined condition of about -1.2 degree / second to about 1.2 degree / second.

[0094] For example, according to the present disclosure, when three basic judgment conditions are satisfied, the process of identifying a short-distance cut-in target can be executed, but the present disclosure is not limited thereto.

[0095] In another example, the system specifications for executing the process of identifying a short-distance cut-in target are as follows.

[0096] First, the short-distance cut-in target can be controlled only when the speed of the vehicle is equal to or less than about 20 kph (about 12.5 mph), but the present disclosure is not limited thereto.

[0097] This is because when the speed of the vehicle is greater than about 20 kph, the effect is negligible and the risk increases in case of failure.

[0098] Second, the target with a relative speed less than about 10 m / s and greater than about -10 m / s can be controlled.

[0099] This is because when the relative speed increases, ODM information delay occurs.

[0100] Third, the moving vehicle moving within about 20 m and the stopped target that has moved previously can be controlled.

[0101] This is because it is impossible to distinguish between a building and a previously stationary stopped vehicle.

[0102] Figure 3 and Figure 4 are diagrams for explaining the process of verifying the validity of ODM information.

[0103] As Figure 3 and Figure 4 shown, when the vehicle satisfies the entry condition for identifying a short-distance cut-in target, the short-distance cut-in target selector according to the present disclosure can verify the validity of the ODM information and can remove the invalid ODM information.

[0104] Here, when verifying the validity of ODM information, the short-distance cut-in target selector can determine the attributes of the ODM information to identify the ODM information from a fixed obstacle, and can determine that the identified ODM information from the fixed obstacle is invalid.

[0105] For example, as Figure 3 shown, when the ODM information is not measured or the absolute speed of the measured ODM information is a value less than the first set value, the short-distance cut-in target selector can determine the attribute of the ODM information as "Not Detected". When the absolute speed of the measured ODM information is a value equal to or greater than the second set value, the short-distance cut-in target selector can determine the attribute of the ODM information as "Moving". When it is determined that the ODM information has the attribute of "Moving" and the absolute speed of the measured ODM information is within the set range, the short-distance cut-in target selector can determine the attribute of the ODM information as "Stopped". When it is not determined that the ODM information has the attribute of "Moving" and the absolute speed of the measured ODM information is a value less than the third set value, the short-distance cut-in target selector can determine the attribute of the ODM information as "Stationary". And when the number of steps is more than a predetermined number of steps and it is determined that the ODM information has the attributes of "Moving" and "Stopped" and has the condition of not being detected, the short-distance cut-in target selector can determine the attribute of the ODM information as "Coasting".

[0106] Here, the short-distance cut-in target selector can set the first set value to about -0.5 m / s, can select and set the second set value from about 0.3 m / s to about 0.7 m / s, can set the set range from more than about -0.5 m / s to the range between about 0.3 m / s and about 0.7 m / s, can select and set the third set value from about 0.3 m / s to about 0.7 m / s, and can set the number of steps to about 3 steps, but the present disclosure is not limited thereto.

[0107] In addition, when the attributes of the ODM information are "Moving", "Stopped", and "Coasting", the short-distance cut-in target selector can determine that the ODM information is valid.

[0108] When the attributes of the ODM information are "Not Detected" and "Stationary", the short-distance cut-in target selector can identify the ODM information as the ODM information from a fixed obstacle, and can determine that the identified ODM information is invalid.

[0109] When necessary, when verifying the validity of ODM information, the short-distance cut-in target selector can determine whether the change in the absolute speed of the ODM information is discontinuous, can identify ODM information from a fixed obstacle, and can determine that the identified ODM information is invalid.

[0110] Here, when the attribute of the ODM information changes from "moving" to "stopped" or from "moving" to "fixed", the short-distance cut-in target selector can determine that the ODM information is invalid.

[0111] For example, when the attribute of the ODM information changes from "moving" to "stopped", the absolute speed of the ODM information can range from a value equal to or greater than about -0.5 m / s to a range between about 0.3 m / s and about 0.7 m / s, and the change in the absolute speed can be greater than about -3 m / s.

[0112] For example, when the attribute of the ODM information changes from "moving" to "fixed", the absolute speed of the ODM information can range from a value equal to or greater than about -0.5 m / s to a range between about 0.3 m / s and about 0.7 m / s, and the change in the absolute speed can be less than about -3 m / s.

[0113] When verifying the validity of the ODM information, the short-distance cut-in target selector can identify the ODM information measured by the target in the vehicle lane in front of the vehicle from the ODM information, and can determine that the identified ODM information is invalid.

[0114] Here, as Figure 4 shown, the short-distance cut-in target selector can determine the attribute of the ODM information existing in the ego-vehicle area 22 occupied by the target 20 in the vehicle lane in front of the vehicle as "not detected".

[0115] The ego-vehicle area 22 occupied by the target 20 in the vehicle lane in front of the vehicle can be calculated by adding the vehicle length and the length margin of the target 20 in the vehicle lane in front of the vehicle.

[0116] For example, the vehicle length of the target 20 in the vehicle lane in front of the vehicle can be about 5 m, and the length margin can be from about 0.5 m to about 1.5 m, but the present disclosure is not limited thereto.

[0117] In this way, according to the present disclosure, a fixed obstacle of the ODM data can be identified through the process of verifying the validity of the ODM data.

[0118] The verification of the ODM data according to the present disclosure needs to meet the following requirements.

[0119] First, it may be necessary to identify ODM information from fixed obstacles such as road surfaces, bridge nodes, small obstacles, or guardrails.

[0120] Second, it may be necessary to identify whether the ODM information is from a moving object or from a stopped object that has previously moved.

[0121] This is because the ODM information is transmitted regardless of the reflecting object.

[0122] Here, the ODM information for identifying a short-distance cut-in can use only the information from the vehicle.

[0123] Then, as Figure 3 shown, according to the present disclosure, the logic for verifying the validity of the ODM data can be designed as follows.

[0124] That is, according to the present disclosure, in order to verify the validity of the ODM data, the attribute of the ODM information can be judged for each area.

[0125] Here, according to the present disclosure, when the attribute of the ODM information is "moving", "stopped", and "coasting", the ODM information can be used as valid information.

[0126] First, when the ODM information is not measured or the measured absolute speed is less than a value of about -0.5 m / s, the attribute of the ODM information can be judged as "not detected".

[0127] Second, when the measured absolute speed is equal to or greater than a value of about 0.3 m / s (to about 0.7 m / s), the attribute of the ODM information can be judged as "moving", and this value is a variable condition regarding the speed of the vehicle itself.

[0128] Third, when the attribute of the ODM information has previously been judged as "moving" and the measured absolute speed is equal to or greater than about -0.5 m / s and less than 0.3 m / s (to about 0.7 m / s), the attribute of the ODM information can be judged as "stopped".

[0129] Fourth, when the attribute of the ODM information has not previously been judged as "moving" and the measured absolute speed is less than a value of about 0.3 m / s (to about 0.7 m / s), the attribute of the ODM information can be judged as "stationary", and this value is a variable condition regarding the speed of the vehicle itself.

[0130] Fifth, when the attribute of the ODM information has been judged as "moving" / "stopped" for more than about three steps (about 150 ms) and the current judgment condition is "not detected", the attribute of the ODM information can be judged as "coasting", and the previous measurement information can be maintained.

[0131] According to the present disclosure, when the change in the absolute speed is identified as discontinuous (i.e., when the vehicle is measured and then a stationary obstacle is measured), the ODM information can be judged as invalid.

[0132] First, the corresponding situation is the case where the attribute of the ODM information changes from "moving" to "stopping".

[0133] For example, the absolute speed is from about -0.5 m / s to about 0.3 m / s (to about 0.7 m / s), and the change in the absolute speed is greater than about -3 m / s, with the change in the negative direction decreasing.

[0134] Second, the corresponding situation is the case where the attribute of the ODM information changes from "moving" to "fixed".

[0135] For example, the absolute speed is from about -0.5 m / s to about 0.3 m / s (to about 0.7 m / s), and the change in the absolute speed is less than about -3 m / s, with the change in the negative direction increasing.

[0136] As Figure 4 shown, according to the present disclosure, the ODM information 24 measured by the forward own-vehicle lane target 20 of the vehicle can be identified through the process of verifying the validity of the ODM data.

[0137] According to the present disclosure, according to the requirements for verifying the validity of the ODM data, it is necessary to identify the ODM information 24 measured by the forward own-vehicle lane target 20 of the vehicle.

[0138] This is because the ODM information 24 from the forward own-vehicle lane target 20 of the vehicle may match the target 40 in the left / right lane, resulting in a misjudgment of a short-distance cut-in.

[0139] Here, the forward own-vehicle lane target 20 is the basic longitudinal control target, so it is not necessary to identify the forward own-vehicle lane target 20 as a short-distance cut-in target.

[0140] Then, according to the present disclosure, the logic for verifying the validity of the ODM data can be designed as follows.

[0141] That is, according to the present disclosure, the ODM information 24 present in the ego vehicle area 22 occupied by the forward own-vehicle lane target 20 can be judged as "not detected".

[0142] The ego vehicle area 22 occupied by the forward own-vehicle lane target 20 can be calculated as follows.

[0143] The area occupied by the forward own-vehicle lane target (Ego Vehicle Area) = vehicle length (e.g., about 5 m) + length margin (e.g., about 0.5 to about 1.5 m).

[0144] Here, the error in the longitudinal position information can be considered to obtain the ODM information 24 of the own vehicle lane target 20 ahead, and here, when the relative speed increases, the error may increase.

[0145] Figure 5 and Figure 6 are diagrams for explaining the process of correcting the ODM information.

[0146] As Figure 5 and Figure 6 shown, the short-distance cut-in target selector according to the present disclosure can correct the ODM information based on the driving state of the own vehicle.

[0147] Here, when correcting the ODM information, the short-distance cut-in target selector can correct the ODM information based on at least one of the offset information and the heading information of the own vehicle.

[0148] That is, when correcting the ODM information, the short-distance cut-in target selector can rearrange the ODM information including the grid map 30 based on the coordinate system of the driving lane.

[0149] As Figure 5 shown, the grid map 32 before correcting the ODM information can be rearranged as the grid map 34 after correcting the ODM information by correcting the ODM data.

[0150] For example, when rearranging the ODM information, the short-distance cut-in target selector can correct the width of the inner region IR of the grid map 30 to be about 0 to about 1.3 m, can correct the width of the first outer region ER1 to be about 1.3 m to about 1.8 m, and can correct the width of the second outer region ER2 of the grid map 30 to be about 1.8 m to about 2.6 m.

[0151] When correcting the width of the inner region IR and the width of the first outer region ER1 of the grid map 30, the short-distance cut-in target selector can apply a margin value of about 0.2 m.

[0152] As described above, the correction of the ODM data according to the present disclosure needs to meet the following requirements.

[0153] First, according to the present disclosure, within the range of satisfying the basic judgment conditions of the short-distance cut-in target, the short-distance cut-in target should not be misidentified due to the behavior of the own vehicle including the own vehicle offset or the own vehicle heading.

[0154] Second, according to the present disclosure, the ODM information needs to be converted and processed into the coordinate system of the driving lane in the own vehicle coordinate system.

[0155] This is because the transmitted ODM information is relative information based on the own vehicle coordinate system.

[0156] According to the present disclosure, the logic for calibrating ODM data can be designed as follows.

[0157] First, the ODM information can be converted into a coordinate system relative to the driving lane by using the longitudinal / lateral position measurement values of each region.

[0158] For example, the vehicle offset and vehicle heading information transmitted from the camera of the present vehicle can be used.

[0159] Second, as Figure 6 shown, the ODM data can be rearranged based on the coordinate system of the driving lane.

[0160] For example, when rearranging the ODM data, the widths of regions IR, ER1, and ER2 can be designed as follows:

[0161] IR_Modified = 0 to 1.3 m, ER1_Modified = 1.3 m to 1.8 m, ER2_Modified = 1.8 m to 2.6 m.

[0162] When judging IR_Modified and ER1_Modified, hysteresis can be applied. For example, the hysteresis can be about 0.2 m.

[0163] Figure 7 is a diagram for explaining the process of selecting a group of ODM track candidates.

[0164] As Figure 7 shown, the short-distance cut-in target selector according to the present disclosure can select a group of ODM track candidates including a plurality of candidate tracks 45 based on the track information.

[0165] Here, when selecting a group of ODM track candidates, the short-distance cut-in target selector can select the tracks in the left lane and the right lane based on the track information as the group of ODM track candidates with the present vehicle lane as the reference.

[0166] That is, when selecting a group of ODM track candidates, the short-distance cut-in target selector can select the tracks existing in a specific distance interval based on the measurements at the longitudinal positions of the left lane and right lane targets as the group of ODM track candidates.

[0167] For example, the specific distance interval can be a distance interval of about -8 m to about 20 m.

[0168] Here, among the tracks existing in the specific distance interval, the tracks corresponding to at most six vehicles in each lane can be selected as the group of ODM track candidates based on the vehicle length of 5 m.

[0169] When selecting the ODM track candidate group, the short-distance cut-in target selector can select candidate track 45 in the distance range of the left and right lanes as the ODM track candidate group, and this distance range is closer to the longitudinal position than the position of the forward vehicle lane target 20 of the vehicle itself.

[0170] Here, the distance range closer to the longitudinal position than the position of the forward vehicle lane target of the vehicle itself can be about -8 m to (longitudinal position of the vehicle in the forward vehicle lane - length margin) m.

[0171] For example, (longitudinal position of the vehicle in the forward vehicle lane - length margin) m can be about 3 m, but it is not limited thereto.

[0172] Then, when measuring the positions of the tracks in the left and right lanes, the short-distance cut-in target selector can set a reference for measuring the positions of the tracks in the left and right lanes to the center of the rear bumper.

[0173] When selecting the ODM track candidate group, the short-distance cut-in target selector can exclude the sensor fusion tracks located at positions farther than the forward vehicle lane target of the vehicle itself from the ODM track candidate group.

[0174] As described above, the selection of the ODM track candidate group according to the present disclosure needs to meet the following requirements.

[0175] First, according to the present disclosure, candidate tracks for determining whether to perform a short-distance cut-in need to be selected from about 64 sensor fusion tracks.

[0176] Second, according to the present disclosure, sensor fusion tracks located at positions farther than the forward vehicle lane target of the vehicle itself need to be excluded from the short-distance cut-in determination candidate group.

[0177] That is, this is because it is only necessary to determine whether to perform a short-distance cut-in for tracks with the possibility of short-distance cut-in.

[0178] According to the present disclosure, the logic for correcting ODM data can be designed as follows.

[0179] First, according to the present disclosure, tracks in the left / right lanes can be selected from about 64 sensor fusion tracks.

[0180] Here, according to the present disclosure, the intermediate information calculated by the target object selection (TOS) module can be utilized, and the tracks can be calculated based on the lateral position of the lane center.

[0181] The tracks calculated by the TOS module can be "moving" tracks or "stopped" tracks that have moved previously.

[0182] Second, according to the present disclosure, when there are tracks in the range of about -8 m to about 20 m based on a reference for measuring the longitudinal position of the left / right lanes, the corresponding tracks can be selected as a candidate group.

[0183] For example, based on a vehicle length of 5 m as a reference, up to six tracks in each lane can be selected as a candidate group.

[0184] Left / right lane tracks closer to the longitudinal position than the position of the forward vehicle lane target 20 can be selected as a candidate group.

[0185] For example, when there are tracks in a specific range based on a reference for measuring the longitudinal position of the left / right lane target, the specific range can be -8 m to {longitudinal position of the ego-lane vehicle - length margin (3 m)}.

[0186] Here, the reference for measuring the position of the track can be the center of the rear bumper.

[0187] Figure 8 and Figure 9 is a view for explaining the process of matching ODM information and candidate tracks with each other.

[0188] As Figure 8 shown, the short-distance cut-in target selector according to the present disclosure can match each candidate track in the candidate tracks 45 and the ODM object 47 corresponding to each candidate track with the grid map 30 of the ODM information.

[0189] When identifying a short-distance cut-in of a candidate track, the short-distance cut-in target selector can add a margin value to the length and width of each of the candidate tracks 45 to set a candidate track matching area 50, and can identify the short-distance cut-in of the candidate tracks 45 based on the set candidate track matching area 50.

[0190] Here, when setting the candidate track matching area 50, the short-distance cut-in target selector can add the length of the candidate track 45 and the length margin to set the longitudinal area of the candidate track matching area 50, and can add the width of the candidate track 45 and the width margin to set the lateral area of the candidate track matching area 50.

[0191] For example, the longitudinal area of the candidate track matching area 50 can be set by adding the length of the candidate track 45, 5 m, and the length margin of about 1 m, and the lateral area of the candidate track matching area 50 can be set by adding the width of the candidate track 45 of about 2 m and the width margin of about 0.8 m.

[0192] When identifying a short-distance cut-in of the candidate track 45, the short-distance cut-in target selector can add a width margin for the heading to the width of each of the candidate tracks 45 to set a candidate track matching area 50, and can identify the short-distance cut-in of the candidate track 45 based on the set candidate track matching area 50.

[0193] Here, when setting the candidate track matching area 50, the short-distance cut-in target selector can multiply the length of the candidate track 45 by the heading angle of the candidate track 45 to calculate a width margin value for the heading, and can add the calculated width margin value for the heading to the width of the candidate track 45 to set the lateral area of the candidate track matching area 50.

[0194] For example, the width margin value for the heading can be set by multiplying the length of the candidate track 45, which is 5m, by the heading angle.

[0195] As Figure 9 shown, when calculating the width margin value for the heading, the short-distance cut-in target selector can also apply a weight to the absolute speed of the target.

[0196] As described above, the matching between the ODM information and the candidate track according to the present disclosure needs to meet the following requirements.

[0197] First, according to the present disclosure, when there is a measurement position including the ODM information of the ODM object 47 in the matching area between the candidate track 45 and the grid map 30, it is necessary to identify a short-distance cut-in.

[0198] Second, according to the present disclosure, when there is no ODM information, it is not necessary to identify a short-distance cut-in.

[0199] Third, according to the present disclosure, after identifying a short-distance cut-in, when the ODM information is unstable, it is also necessary to maintain the identification of the short-distance cut-in target.

[0200] Fourth, according to the present disclosure, after identifying a short-distance cut-in, when the information of the candidate track is unstable, it is also necessary to maintain the identification of the short-distance cut-in target.

[0201] According to the present disclosure, the logic for matching the ODM information and the candidate track can be designed as follows.

[0202] First, according to the present disclosure, a margin can be added to the length and width of the candidate track 45 to set the candidate track matching area 50, which can be performed based on the measurement position of the candidate track.

[0203] That is, the longitudinal area of the candidate track 45 can be set by adding the length (5m) and the length margin (1.0m).

[0204] Additionally, the lateral area of the candidate orbit 45 can be set by adding the width (2m) and the width margin (0.8m).

[0205] Second, a width margin for the heading of the candidate orbit can be added to set the candidate orbit matching area 50, and in this case, the width margin for the heading can be applied only when the width margin for the heading is greater than the general width margin.

[0206] That is, the width margin for the heading can be calculated as follows.

[0207] Width margin (HA) = length (5m) × sin (orbit heading angle).

[0208] Here, as Figure 9 shown, when a width margin for the heading of the candidate orbit is added to set the candidate orbit matching area 50, a weight can also be applied to the absolute speed of the target.

[0209] This is because the incidence of misidentification needs to be reduced.

[0210] Figure 10 is a diagram for explaining the process of identifying a short - distance cut - in of a candidate orbit.

[0211] As Figure 10 shown, according to the short - distance cut - in target selector of the present disclosure, each candidate orbit and the ODM object corresponding to each candidate orbit can be matched with a grid map of ODM information to identify the short - distance cut - in of the candidate orbit.

[0212] That is, when identifying the short - distance cut - in of a candidate orbit, the short - distance cut - in target selector can match each candidate orbit and the ODM object corresponding to each candidate orbit with the grid map, and can judge the short - distance cut - in state of each candidate orbit based on the match.

[0213] Here, when judging the short - distance cut - in state of each candidate orbit, the short - distance cut - in target selector can judge the short - distance cut - in state of the candidate orbit as one of the undetected state 510, new state 520, updated state 530, orbit inertial coasting state 550, ODM inertial coasting state 540, and time inertial coasting state 560 based on the match.

[0214] For example, when there is no ODM information or candidate orbit, the short - distance cut - in target selector can judge the short - distance cut - in state of the candidate orbit as the undetected state 510.

[0215] Here, when the candidate orbit and the ODM object corresponding to the candidate orbit match the internal area IR of the grid map, as Figure 10As shown in (1) below, the short-distance cut-in target selector can change the short-distance cut-in state from the undetected state 510 to the new state 520, and can determine the short-distance cut-in state of the candidate orbit.

[0216] When the orbit identified for a general cut-in is the same as the candidate orbit and the candidate orbit matches the first external region ER1 of the grid map, as Figure 10 shown in (15) below, the short-distance cut-in target selector can change the short-distance cut-in state from the undetected state 510 to the updated state 530, and determine the short-distance cut-in state of the candidate orbit.

[0217] Then, when the candidate orbit and the ODM object corresponding to the candidate orbit match the internal region IR of the grid map, the short-distance cut-in target selector can determine the short-distance cut-in state of the candidate orbit as the new state 520.

[0218] Here, when the new state is maintained for a predetermined time or a predetermined number of steps, as Figure 10 shown in (2) below, the short-distance cut-in target selector can change from the new state 520 to the updated state 530, and can determine the short-distance cut-in state of the candidate orbit.

[0219] For example, when the new state 520 is maintained for about 0.25 s or about 5 steps, the short-distance cut-in target selector can change from the new state 520 to the updated state 530.

[0220] When the match with the internal region IR of the grid map is released before the new state 520 is maintained for a predetermined time or a predetermined number of steps, as Figure 10 shown in (3) below, the short-distance cut-in target selector can change from the new state 520 to the undetected state 510, and can determine the short-distance cut-in state of the candidate orbit.

[0221] For example, when the match with the internal region IR of the grid map is released before the new state 520 is maintained for about 0.25 s or about 5 steps, as Figure 10 shown in (3) below, the short-distance cut-in target selector can change from the new state 520 to the undetected state 510.

[0222] Then, when the candidate orbit and the ODM object corresponding to the candidate orbit match the internal region IR or the first external region ER1 of the grid map, the short-distance cut-in target selector can identify the short-distance cut-in of the candidate orbit as the updated state.

[0223] When the ODM object corresponding to the candidate orbit disappears and the match with the grid map is released, as Figure 10As shown in (4), the short-distance cut-in target selector can transition from the update state 530 to the orbital inertial glide state 550, and can determine the short-distance cut-in state of the candidate orbit.

[0224] When the candidate orbit disappears and the matching with the grid map is released while the host vehicle stops and the candidate orbit stops, as Figure 10 shown in (5), the short-distance cut-in target selector can transition from the update state 530 to the ODM inertial glide state 540, and can determine the short-distance cut-in state of the candidate orbit.

[0225] When the candidate orbit disappears and the matching with the grid map is released while the host vehicle is moving or the candidate orbit is moving, as Figure 10 shown in (6), the short-distance cut-in target selector can transition from the update state 530 to the undetected state 510, and can determine the short-distance cut-in state of the candidate orbit.

[0226] Then, when the candidate orbit and the ODM object corresponding to the candidate orbit both disappear and the matching with the grid map is released while the host vehicle stops or the candidate orbit stops, as Figure 10 shown in (7), the short-distance cut-in target selector can transition from the update state 530 to the time inertial glide state 560, and can determine the short-distance cut-in state of the candidate orbit.

[0227] Then, when the ODM object corresponding to the candidate orbit disappears and the matching with the grid map is released, the short-distance cut-in target selector can determine the short-distance cut-in state of the candidate orbit as the orbital inertial glide state 550.

[0228] When the ODM object corresponding to the candidate orbit is restored and matches the internal region IR or the first external region ER1 of the grid map, as Figure 10 shown in (8), the short-distance cut-in target selector can transition from the orbital inertial glide state 550 to the update state 530, and can determine the short-distance cut-in state of the candidate orbit.

[0229] For example, when the matching between the ODM object corresponding to the candidate orbit and the grid map is maintained for a predetermined time or a predetermined number of steps, as Figure 10 shown in (8), the short-distance cut-in target selector can transition from the orbital inertial glide state 550 to the update state 530.

[0230] Here, the short-distance cut-in target selector can maintain the matching between the ODM object corresponding to the candidate orbit and the grid map for about 0.15 s or about three steps, but the present disclosure is not limited thereto.

[0231] When the candidate orbit disappears while the host vehicle stops and the candidate orbit stops, as Figure 10As shown in (9), the short-distance cut-in target selector can transition from the orbital inertial coasting state 550 to the time inertial coasting state 560 and can determine the short-distance cut-in state of the candidate orbit.

[0232] When the candidate orbit disappears and the match with the grid map is released while the vehicle is moving or the candidate orbit is moving, as Figure 10 shown in (14), the short-distance cut-in target selector can transition from the orbital inertial coasting state 550 to the undetected state 510 and can determine the short-distance cut-in state of the candidate orbit.

[0233] Then, when the candidate orbit disappears and the match with the grid map is released while the vehicle is stopped and the candidate orbit is stopped, the short-distance cut-in target selector can determine the short-distance cut-in state of the candidate orbit as the ODM inertial coasting state 540.

[0234] Here, when the candidate orbit is restored and matches the internal region IR or the first external region ER1 of the candidate orbit, as Figure 10 shown in (10), the short-distance cut-in target selector can transition from the ODM inertial coasting state 540 to the updated state 530 and can determine the short-distance cut-in state of the candidate orbit.

[0235] For example, when the match between the candidate orbit and the grid map is maintained for a predetermined time or a predetermined number of steps, as Figure 10 shown in (10), the short-distance cut-in target selector can transition from the ODM inertial coasting state 540 to the updated state 530.

[0236] Here, the short-distance cut-in target selector can maintain the match between the candidate orbit and the grid map for about 0.15 s or about 3 steps, but the present disclosure is not limited thereto.

[0237] When the ODM object corresponding to the candidate orbit changes rapidly, the ODM object corresponding to the candidate orbit disappears, or the vehicle moves independently of time, as Figure 10 shown in (11), the short-distance cut-in target selector can transition from the ODM inertial coasting state 540 to the undetected state 510 and can determine the short-distance cut-in state of the candidate orbit.

[0238] Here, when the ODM object corresponding to the candidate orbit changes rapidly before about 1 s after entry or the ODM object corresponding to the candidate orbit disappears after about 1 s after entry, as Figure 10 shown in (11), the short-distance cut-in target selector can transition from the ODM inertial coasting state 540 to the undetected state 510 and can determine the short-distance cut-in state of the candidate orbit.

[0239] Then, when the candidate track and the ODM object corresponding to the candidate track both disappear and release the match with the grid map when the vehicle stops or the candidate track stops, the short-distance cut-in target selector can determine the short-distance cut-in state of the candidate track as the time inertial gliding state 560.

[0240] Here, when the candidate track and the ODM object corresponding to the candidate track both recover and match the internal region IR or the first external region ER1 of the grid map, as Figure 10 shown in (12) of, the short-distance cut-in target selector can change from the time inertial gliding state 560 to the update state 530, and can determine the short-distance cut-in state of the candidate track.

[0241] For example, when the match between the candidate track and the ODM object corresponding to the candidate track and the grid map is maintained for a predetermined time or a predetermined number of steps, as Figure 10 shown in (12) of, the short-distance cut-in target selector can change from the time inertial gliding state 560 to the update state 530.

[0242] Here, the short-distance cut-in target selector can maintain the match between the candidate track and the ODM object corresponding to the candidate track for about 0.15 s or about three steps, but the present disclosure is not limited thereto.

[0243] When the match with the grid map is maintained for a predetermined time or a predetermined number of steps in the time inertial gliding state 560 or the vehicle moves, as Figure 10 shown in (13) of, the short-distance cut-in target selector can change from the time inertial gliding state 560 to the undetected state 510, and can determine the short-distance cut-in state of the candidate track.

[0244] Here, the short-distance cut-in target selector can maintain the match with the grid map for about 1 s or about 20 steps in the time inertial gliding state, but the present disclosure is not limited thereto.

[0245] Figure 11 is a diagram for explaining the process of determining the short-distance cut-in target.

[0246] As Figure 11 shown, the short-distance cut-in target selector according to the present disclosure can determine the short-distance cut-in target from the short-distance cut-in candidate track, and can output the determined short-distance cut-in target information.

[0247] Here, when identifying the short-distance cut-in target, the short-distance cut-in target selector can determine the candidate track closest to the vehicle 10 among the candidate tracks 45 for short-distance cut-in identification in the left lane and the right lane based on the vehicle's lane as the short-distance cut-in target.

[0248] For example, when determining a short-distance cut-in target, the short-distance cut-in target selector may select the first candidate track 45-1 closest to the own vehicle 10 among the candidate tracks 45 identified for short-distance cut-in in the left lane based on the own vehicle lane, may select the second candidate track 45-2 closest to the own vehicle 10 among the candidate tracks 45 identified for short-distance cut-in in the right lane based on the own vehicle lane, and may determine the two candidate tracks including the first candidate track 45-1 and the second candidate track 45-2 as the short-distance cut-in target.

[0249] When outputting the determined short-distance cut-in target information, the short-distance cut-in target selector may calculate position information and relative speed information based on the measurement points of the short-distance cut-in target, and may output the short-distance cut-in target information including the calculated position information and relative speed information.

[0250] As described above, the determination of the short-distance cut-in target and the transmission of information according to the present disclosure need to meet the following requirements.

[0251] First, according to the present disclosure, the closest track among the tracks identified for short-distance cut-in in the left / right lane may be selected as the short-distance cut-in target.

[0252] Second, according to the present disclosure, each short-distance cut-in target may be selected in the left / right lane, and the two target information may be transmitted to the controller.

[0253] Third, according to the present disclosure, position information and relative speed information may be calculated using the target information based on the measurement points of the track.

[0254] According to the present disclosure, a short-distance cut-in target may be determined, and the logic for transmitting information may be designed as follows.

[0255] First, according to the present disclosure, when determining a short-distance cut-in target, the closest candidate track may be selected from the candidate tracks identified as "detected".

[0256] Second, the information transmitted to the controller may be as follows:

[0257] LEFT / RIGHTCloseCutInTgt_Status_L / R: Status of the left / right target;

[0258] 0x0: Not detected;

[0259] 0x1: New;

[0260] 0x2: Updated;

[0261] 0x3: Inertial coasting (ODM disappeared);

[0262] 0x4: Coasting (track disappears);

[0263] 0x5: Coasting (ODM disappears and track disappears, time coasting);

[0264] CloseCutInTgt_RelPos_L / R: Relative distance to the left / right target, mps (transmitting track information);

[0265] CloseCutInTgt_RelSpd_L / R: Relative speed to the left / right target, mps (transmitting track information).

[0266] Figures 12 to 14 It is a diagram for explaining the application of the short - distance cut - in target recognition device according to an embodiment of the present disclosure.

[0267] Figure 12 It is a diagram for explaining the first scenario of recognizing short - distance cut - in vehicles in the left / right lanes while controlling the stop of the vehicle.

[0268] As Figure 12 shown, as the first operation, the host vehicle 10 can stop the leading vehicle 60 while performing track control on the leading vehicle 60.

[0269] Then, as the second operation, the host vehicle 10 can perform stop control at an interval of about 2.5 m to about 3.5 m from the leading vehicle 60.

[0270] Then, as the third operation, the short - distance cut - in target recognition device of the host vehicle 10 can recognize the short - distance cut - in vehicle 70.

[0271] In addition, as the fourth operation, when the leading vehicle 60 starts, the host vehicle 10 can remain stopped after the leading vehicle 60 starts, thereby preventing a collision with the short - distance cut - in vehicle 70.

[0272] Here, in the holding state, even if a start command (operation on the + / - RES switch and the accelerator pedal) is received, the host vehicle 10 can be controlled to prevent starting.

[0273] Figure 13 It is a diagram for explaining the second scenario of recognizing short - distance cut - in vehicles in the left / right lanes when the leading vehicle moves and then stops during the stop control of the host vehicle.

[0274] As Figure 13 shown, as the first operation, the host vehicle 10 can stop the leading vehicle 60 while performing track control on the leading vehicle 60.

[0275] Then, as a second operation, the host vehicle 10 can perform stop control at an interval of about 2.5 m to about 3.5 m from the preceding vehicle 60.

[0276] Then, as a third operation, the host vehicle 10 can enter a holding state and then can remain stopped after the preceding vehicle 60 starts.

[0277] As a fourth operation, the short-distance cut-in target recognition device of the host vehicle 10 can recognize the short-distance cut-in vehicle 70.

[0278] Then, as a fifth operation, the host vehicle 10 can start in response to a start command (operation of the + / -RES switch and the accelerator pedal) and can perform stop control at an interval of about 2.5 m to about 3.5 m from the recognized short-distance cut-in vehicle 70.

[0279] Figure 14 is a diagram for explaining a third scenario in which a short-distance cut-in vehicle in the left / right lane is recognized while the host vehicle performs trajectory control on the preceding vehicle.

[0280] As Figure 14 shown, as a first operation, the preceding vehicle 60 can travel at a speed of about 20 kph or less.

[0281] Then, as a second operation, the host vehicle 10 can perform trajectory control of the preceding vehicle 60.

[0282] Then, as a third operation, the short-distance cut-in target recognition device of the host vehicle 10 can recognize the short-distance cut-in vehicle 70.

[0283] In addition, the host vehicle 10 can perform acceleration / deceleration control on the recognized short-distance cut-in vehicle 70.

[0284] As described above, according to the present disclosure, various situations of an oncoming vehicle that cuts in at a low speed within a short distance can be accurately determined, thereby improving reliability and safety.

[0285] Figure 15 is a flowchart for explaining a method of recognizing a short-distance cut-in target according to the present disclosure.

[0286] As Figure 15 shown, according to the present disclosure, it can be determined whether the host vehicle satisfies the short-distance cut-in target recognition entry condition (S10).

[0287] Here, according to the present disclosure, the driving state information of the host vehicle including the offset information, heading information, and turning information of the host vehicle can be received, and it can be determined whether the driving state of the host vehicle satisfies the short-distance cut-in target recognition entry condition based on the driving state information of the host vehicle.

[0288] In this case, according to the present disclosure, it is possible to determine whether the driving state of the vehicle satisfies at least one of the offset driving condition, the heading driving condition, and the turning driving condition.

[0289] According to the present disclosure, in the case of a short-distance cut-in to the target recognition entry condition, the validity of the ODM information can be verified (S20).

[0290] Here, according to the present disclosure, as a verification result of the validity of the ODM information, when the ODM information is invalid, the invalid ODM information can be removed (S30).

[0291] For example, according to the present disclosure, it is possible to determine the attribute of the ODM information to identify the ODM information from a fixed obstacle, and the identified ODM information can be determined to be invalid.

[0292] If necessary, according to the present disclosure, it is possible to identify the ODM information measured by the target of the vehicle lane in front from the ODM information, and the identified ODM information can be identified as invalid.

[0293] Then, according to the present disclosure, the ODM information can be corrected based on the driving state of the vehicle (S40).

[0294] Here, according to the present disclosure, the ODM information can be corrected based on at least one of the offset information and the heading information of the vehicle.

[0295] In this case, according to the present disclosure, the ODM information including the grid map can be rearranged based on the coordinate system of the driving lane.

[0296] Then, according to the present disclosure, an ODM track candidate group including a plurality of candidate tracks can be selected based on the track information (S50).

[0297] Here, according to the present disclosure, tracks in the left lane and the right lane based on the vehicle lane can be selected from a plurality of sensor fusion tracks as the ODM track candidate group based on the track information.

[0298] According to the present disclosure, tracks existing in a specific distance interval based on the reference for measuring the longitudinal positions of the left lane and the right lane targets can be selected as the ODM track candidate group.

[0299] According to the present disclosure, tracks in the left lane and the right lane in a distance interval closer to the longitudinal position than the position of the vehicle lane target in front can be selected as the ODM track candidate group.

[0300] According to the present disclosure, each candidate track and the ODM object corresponding to each candidate track can be matched with the grid map of the ODM information to identify a short-distance cut-in of the candidate track (S60).

[0301] Here, according to the present disclosure, a margin value can be added to the length and width of each candidate track to set a candidate track matching region, and a short-distance cut-in of the candidate track can be identified based on the set candidate track matching region.

[0302] For example, according to the present disclosure, the length of the candidate track and the length margin can be added to set the longitudinal region of the candidate track matching region, and the width of the candidate track and the width margin can be added to set the lateral region of the candidate track matching region.

[0303] If necessary, according to the present disclosure, a width margin value for the course can be added to the width of each candidate track to set a candidate track matching region, and a short-distance cut-in of the candidate track can also be identified based on the set candidate track matching region.

[0304] According to the present disclosure, based on the matching, the short-distance cut-in state of the candidate track can be determined as one of an undetected state, a new state, an updated state, a track inertial coasting state, an ODM inertial coasting state, and a time inertial coasting state.

[0305] Then, according to the present disclosure, a short-distance cut-in target can be determined among the short-distance cut-in candidate tracks (S70).

[0306] Here, according to the present disclosure, among the candidate tracks identified as short-distance cut-in tracks in the left lane and the right lane based on the own vehicle lane, the candidate track closest to the own vehicle can be identified as the short-distance cut-in target.

[0307] For example, according to the present disclosure, a first candidate track closest to the own vehicle among the candidate tracks identified as short-distance cut-in tracks in the left lane based on the own vehicle lane can be selected, a second candidate track closest to the own vehicle among the candidate tracks identified as short-distance cut-in tracks in the right lane based on the own vehicle lane can be selected, and the two candidate tracks including the first candidate track and the second candidate track can be determined as the short-distance cut-in target.

[0308] Then, according to the present disclosure, the determined short-distance cut-in target information can be output (S80).

[0309] Here, according to the present disclosure, position information and relative speed information can be calculated based on the measurement point of the short-distance cut-in target, and the short-distance cut-in target information including the calculated position information and relative speed information can be output.

[0310] According to the present disclosure, it can be checked whether the short-distance cut-in target identification is terminated (S90), and when the short-distance cut-in target identification is terminated, the short-distance cut-in target identification process can be terminated.

[0311] According to the present disclosure, a computer-readable recording medium storing a program for executing an identification method of a short-distance cut-in target recognition device may execute the processes provided in the method for identifying a short-distance cut-in target.

[0312] A vehicle according to an embodiment of the present disclosure may include: a sensing device configured to sense the host vehicle and surrounding objects; and a short-distance cut-in target recognition device configured to identify a short-distance cut-in vehicle based on the host vehicle and surrounding object information received from the sensing device, wherein the short-distance cut-in target recognition device includes: an ODM information calculator configured to calculate ODM information based on the host vehicle and surrounding object information; an orbit information calculator configured to calculate orbit information based on the host vehicle and surrounding object information; and a short-distance cut-in target selector configured to determine whether the host vehicle satisfies the short-distance cut-in target recognition entry condition, verify the validity of the ODM information and remove invalid ODM information when the host vehicle satisfies the short-distance cut-in target recognition entry condition, correct the ODM information based on the driving state of the host vehicle, select an ODM orbit candidate group including a plurality of candidate orbits based on the orbit information, match each candidate orbit and the ODM object corresponding to each candidate orbit with a grid map of the ODM information to identify the short-distance cut-in of the candidate orbit, determine a short-distance cut-in target from the short-distance cut-in candidate orbits, and output the determined short-distance cut-in target information.

[0313] As described above, according to the present disclosure, the ODM information corrected based on the driving state of the host vehicle and the plurality of candidate orbits selected based on the orbit information may be matched with each other to identify a short-distance cut-in target vehicle, thereby improving reliability and safety.

[0314] Therefore, according to the present disclosure, various situations of an oncoming vehicle cutting in at a low speed within a short distance can be accurately determined, thereby improving reliability and safety.

[0315] In the short-distance cut-in target recognition device and its recognition method related to at least one embodiment of the present disclosure configured as described above, the ODM information corrected based on the driving state of the host vehicle and the plurality of candidate orbits selected based on the orbit information may be matched with each other to identify a short-distance cut-in target vehicle, thereby improving reliability and safety.

[0316] Therefore, according to the present disclosure, various situations of an oncoming vehicle cutting in at a low speed within a short distance can be accurately determined, thereby improving reliability and safety.

[0317] Those skilled in the art will recognize that the effects that the present disclosure can achieve are not limited to the specific content described above, and other advantages of the present disclosure will be more clearly understood from the detailed description.

[0318] The foregoing disclosure may also be implemented as computer-readable code stored in a computer-readable recording medium. A computer-readable recording medium is any data storage device that can store data that can be subsequently read by a computer. Examples of computer-readable recording media include hard disk drives (HDDs), solid state drives (SSDs), silicon disk drives (SDDs), read-only memories (ROMs), random access memories (RAMs), CD-ROMs, magnetic tapes, floppy disks, optical data storage devices, carrier waves (e.g., transmission via the Internet), and the like.

[0319] It will be apparent to those skilled in the art that various modifications and changes can be made to the disclosure without departing from the spirit or scope of the embodiments. Accordingly, if the modifications and changes of the embodiments fall within the scope of the appended claims and their equivalents, the disclosure is intended to cover such modifications and changes.

Claims

1. A short-distance cut-in target recognition device, comprising: An Occupancy Distance Map information calculator, i.e., an ODM information calculator, which calculates ODM information based on the information of the vehicle itself and surrounding objects; A trajectory information calculator, which calculates trajectory information based on the information of the vehicle itself and surrounding objects; And A short-distance cut-in target selector, which selects a short-distance cut-in target based on the ODM information and the trajectory information, wherein the short-distance cut-in target selector determines whether the vehicle itself meets the short-distance cut-in target recognition entry condition. When the vehicle itself meets the short-distance cut-in target recognition entry condition, it verifies the validity of the ODM information and removes invalid ODM information, corrects the ODM information based on the driving state of the vehicle itself, selects an ODM trajectory candidate group including multiple candidate trajectories based on the trajectory information, matches each candidate trajectory in the multiple candidate trajectories and the ODM object corresponding to each candidate trajectory with the grid map of the ODM information to identify the short-distance cut-in of the candidate trajectory, determines a short-distance cut-in target from the short-distance cut-in candidate trajectories, and outputs the determined short-distance cut-in target information.

2. The short-distance cut-in target recognition device according to claim 1, wherein, When determining whether the vehicle itself meets the short-distance cut-in target recognition entry condition, the short-distance cut-in target selector receives the driving state information of the vehicle itself, and the driving state information includes the offset information, heading information and turning information of the vehicle itself. The short-distance cut-in target selector determines whether the driving state of the vehicle itself meets the short-distance cut-in target recognition entry condition based on the driving state information of the vehicle itself.

3. The short-distance cut-in target recognition device according to claim 1, wherein, When verifying the validity of the ODM information, the short-distance cut-in target selector determines the attribute of the ODM information to identify the ODM information from a fixed obstacle, and determines that the identified ODM information is invalid.

4. The short-distance cut-in target recognition device according to claim 1, wherein, When verifying the validity of the ODM information, the short-distance cut-in target selector identifies the ODM information measured by the target of the vehicle lane in front from the ODM information, and determines that the identified ODM information is invalid.

5. The short-distance cut-in target recognition device according to claim 1, wherein, When correcting the ODM information, the short-distance cut-in target selector corrects the ODM information based on the offset information or heading information of the vehicle itself.

6. The short-distance cut-in target recognition device according to claim 1, wherein, When selecting the ODM trajectory candidate group, the short-distance cut-in target selector selects trajectories in the left lane and the right lane based on the vehicle lane from multiple sensor fusion trajectories as the ODM trajectory candidate group based on the trajectory information.

7. The short-distance cut-in target recognition device according to claim 1, wherein, When identifying a short-distance cut-in of the candidate track, the short-distance cut-in target selector adds a margin value to the length and width of each candidate track to set a candidate track matching area, and identifies the short-distance cut-in of the candidate track based on the set candidate track matching area.

8. The short-distance cut-in target recognition device according to claim 1, wherein When identifying a short-distance cut-in of the candidate track, the short-distance cut-in target selector adds a width margin value for the course to the width of each candidate track to set a candidate track matching area, and identifies the short-distance cut-in of the candidate track based on the set candidate track matching area.

9. The short-distance cut-in target recognition device according to claim 1, wherein When determining the short-distance cut-in target, the short-distance cut-in target selector determines, as the short-distance cut-in target, the candidate track closest to the host vehicle among the candidate tracks identified for short-distance cut-in in the left and right lanes with the host vehicle lane as a reference.

10. The short-distance cut-in target recognition device according to claim 1, wherein When outputting the determined short-distance cut-in target information, the short-distance cut-in target selector calculates position information and relative speed information based on the measurement point of the short-distance cut-in target, and outputs short-distance cut-in target information including the calculated position information and relative speed information.

11. A method for identifying a short-distance cut-in target based on ODM information and track information, the method comprising: Judging whether the host vehicle meets the short-distance cut-in target recognition entry condition; When the host vehicle meets the short-distance cut-in target recognition entry condition, verifying the validity of the ODM information and removing invalid ODM information; Correcting the ODM information based on the driving state of the host vehicle; Selecting an ODM track candidate group including a plurality of candidate tracks based on the track information; Matching each candidate track among the plurality of candidate tracks and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify the short-distance cut-in of the candidate track; Determining a short-distance cut-in target from the short-distance cut-in candidate tracks; And Outputting the determined short-distance cut-in target information.

12. The method according to claim 11, wherein Judging whether the host vehicle meets the short-distance cut-in target recognition entry condition includes: Receiving the driving state information of the host vehicle, the driving state information including the offset information, course information and turning information of the host vehicle; and Judging whether the driving state of the host vehicle meets the short-distance cut-in target recognition entry condition based on the driving state information of the host vehicle.

13. The method according to claim 11, wherein Identifying the validity of the ODM information includes: Judging the attribute of the ODM information; Identifying the ODM information from fixed obstacles; and Judging that the identified ODM information is invalid.

14. The method according to claim 11, wherein Identifying the validity of the ODM information includes: Identify the ODM information measured by the target in the lane of the host vehicle ahead from the ODM information; and Determine that the identified ODM information is invalid.

15. The method according to claim 11, wherein Calibrating the ODM information includes: Calibrating the ODM information based on the offset information or heading information of the host vehicle.

16. The method according to claim 11, wherein Selecting the ODM track candidate group includes: Selecting tracks in the left lane and the right lane based on the host vehicle lane from multiple sensor fusion tracks as the ODM track candidate group based on the track information.

17. The method according to claim 11, wherein Determining the short-distance cut-in target includes: Determining the candidate track closest to the host vehicle among the candidate tracks identified for short-distance cut-in in the left lane and the right lane based on the host vehicle lane as the short-distance cut-in target.

18. The method according to claim 11, wherein Outputting the determined short-distance cut-in target information includes: Calculating position information and relative speed information based on the measurement point of the short-distance cut-in target; and Outputting the short-distance cut-in target information including the calculated position information and relative speed information.

19. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method according to claim 11.

20. A vehicle, comprising: A sensing device that senses the host vehicle and surrounding objects; And A short-distance cut-in target recognition device that recognizes a short-distance cut-in vehicle based on the information of the host vehicle and surrounding objects received from the sensing device, Wherein the short-distance cut-in target recognition device includes: An occupancy distance map information calculator, i.e., an ODM information calculator, that calculates ODM information based on the information of the host vehicle and surrounding objects; A track information calculator that calculates track information based on the information of the host vehicle and surrounding objects; and A short-distance cut-in target selector that determines whether the host vehicle meets the short-distance cut-in target recognition entry condition, validates the effectiveness of the ODM information and removes invalid ODM information when the host vehicle meets the short-distance cut-in target recognition entry condition, calibrates the ODM information based on the driving state of the host vehicle, selects an ODM track candidate group including multiple candidate tracks based on the track information, matches each candidate track in the multiple candidate tracks and the ODM object corresponding to each candidate track with the grid map of the ODM information to identify the short-distance cut-in of the candidate track, determines the short-distance cut-in target from the short-distance cut-in candidate tracks, and outputs the determined short-distance cut-in target information.

Images