Autonomous driving vehicle and method for operating same

Abstract

Disclosed are an autonomous vehicle and a method for operating the same, the autonomous vehicle including: at least one sensor configured to detect an ambient environment of the vehicle and generate ambient environment information; a processor configured to monitor a state of the vehicle to generate vehicle state information, and determine whether a minimum risk maneuver is required based on at least one of the surroundings information or the vehicle state information during autonomous driving of the vehicle; and a controller configured to control an operation of the vehicle according to control of the processor, and the processor may be configured to: set a region of interest when it is determined that a minimum risk maneuver is required and it is determined that the road shoulder parking type is a type of the minimum risk maneuver; searching for a potential MRC region, the potential MRC region being a candidate MRC region for parking on a road shoulder within the region of interest; and selecting, among the potential MRC regions, a target MRC region in which the vehicle is to be parked.

Description

[0001] Cross - reference to related applications

[0002] This application claims the priority and benefit of Korean Patent Application No. 10 - 2024 - 0015855, filed on February 1, 2024, the entire contents of which are incorporated herein by reference as part of the disclosure of this patent document. Technical field

[0003] Various embodiments of the disclosed technology relate to a vehicle configured to search for and select a position for a shoulder stop during autonomous driving and a method for operating the vehicle. Background art

[0004] Recently, advanced driver assistance systems (ADAS) have been developed to assist drivers in driving. ADAS has multiple sub - categories and provides convenience for drivers. Such ADAS is also referred to as autonomous driving or ADS (automatic driving system).

[0005] Meanwhile, an abnormality may occur in the autonomous driving system while the vehicle is performing autonomous driving. If no appropriate measures are taken against such an abnormality in the autonomous driving system, the vehicle may enter a dangerous state. Summary of the invention

[0006] Accordingly, various embodiments of the present disclosure disclose a vehicle configured to search for and select a position for a shoulder stop as a type of minimum - risk maneuver during autonomous driving when a shoulder stop is required.

[0007] Various embodiments of the present disclosure disclose a method for operating a vehicle configured to search for and select a position to stop on the shoulder of a road as a type of minimum - risk maneuver during autonomous driving when a shoulder stop is required.

[0008] The technical objects achieved by the present disclosure are not limited to the foregoing objects, and other technical objects will be apparent to those skilled in the art to which the present disclosure pertains from the following description.

[0009] One embodiment is an autonomous vehicle, comprising: at least one sensor configured to detect the surrounding environment of the vehicle and generate surrounding environment information; a processor configured to monitor the state of the vehicle to generate vehicle state information, and determine whether a minimum risk maneuver is required based on at least one of the surrounding environment information or the vehicle state information during autonomous driving of the vehicle; and a controller configured to control the operation of the vehicle according to the control of the processor, and the processor may be configured to: set a region of interest when it is determined that a minimum risk maneuver is required and it is determined that the type of shoulder parking is the type of minimum risk maneuver; search for potential MRC regions, which are candidate MRC regions for shoulder parking within the region of interest; and select a target MRC region where the vehicle will park among the potential MRC regions.

[0010] The region of interest may have a rectangular shape including an upper side, a lower side, a left side, and a right side. The upper side may have a length from a point at the front of the vehicle to a position separated from the front of the vehicle by a predetermined first distance, and the left side may have a length from a point on the center line to the boundary of the shoulder.

[0011] The processor may search for potential MRC regions based on at least one of pre-stored map information, surrounding environment information, and traffic information.

[0012] The processor may determine, within the region of interest, a region that is unoccupied by obstacles, greater than a predetermined reference size, and where the vehicle can stop by decelerating and moving from the current position of the vehicle to a different lane as a potential MRC region.

[0013] The processor may search for potential MRC regions for a predetermined time.

[0014] After the search for potential MRC regions is completed, the processor may continue to search for potential MRC regions in the background.

[0015] The potential MRC regions may include a plurality of candidate MRC regions, and each of the candidate MRC regions may have a different size.

[0016] A gap region may be formed between a candidate MRC region near an obstacle and the obstacle among the plurality of candidate MRC regions.

[0017] Based on an obstacle located on the road shoulder, the potential MRC area may include: a first candidate MRC area that is located in the forward direction of the obstacle as the direction of vehicle approach; and a second candidate MRC area that is located in the backward direction of the obstacle as the direction away from the vehicle based on the obstacle. A first gap area may be formed between the first candidate MRC area and the obstacle, and a second gap area may be formed between the obstacle and the second candidate MRC area.

[0018] The second gap area may be larger than the first gap area.

[0019] The processor may select a target MRC area in the potential MRC area based on at least one of information about the size of the potential MRC area, information about the distance to the potential MRC area, information about the movement path to the potential MRC area, information about the vehicle state, information about the number of lane changes, surrounding environment information, and traffic information.

[0020] When selecting the target MRC area, the processor may control the vehicle to move into the target MRC area.

[0021] Another embodiment is a method for operating an autonomous vehicle, including: setting an area of interest when it is determined during the autonomous driving of the vehicle that a minimum risk maneuver is required based on at least one of surrounding environment information or vehicle state information and the road shoulder parking type is determined to be the type of minimum risk maneuver; searching for a potential MRC area that is a candidate MRC area for road shoulder parking within the area of interest; and selecting a target MRC area where the vehicle will park among the potential MRC areas.

[0022] Searching for the potential MRC area may include searching for the potential MRC area based on at least one of pre-stored map information, surrounding environment information, and traffic information.

[0023] Searching for the potential MRC area may include: within the area of interest, determining an area that the vehicle can stop by decelerating and moving from the current position of the vehicle to a different lane, is not occupied by an obstacle, and is larger than a predetermined reference size as the potential MRC area.

[0024] Searching for the potential MRC area may include searching for the potential MRC area for a predetermined time.

[0025] Searching for the potential MRC area may include: after completing the search for the potential MRC area, continuing to search for the potential MRC area in the background.

[0026] Selecting a target MRC region may include: selecting a target MRC region from potential MRC regions based on at least one of information about the size of a potential MRC region, information about the distance to a potential MRC region, information about the movement path to a potential MRC region, information about the vehicle state, information about the number of lane changes, surrounding environment information, and traffic information.

[0027] The method may further include: when selecting a target MRC region, controlling the vehicle to move into the target MRC region. Description of the Drawings

[0028] Figure 1 is a block diagram of a vehicle according to various embodiments of the present disclosure.

[0029] Figure 2 is a functional block diagram of a processor showing according to various embodiments of the present disclosure.

[0030] Figure 3 is a diagram showing a minimum risk maneuver (MRM) strategy according to vehicle state according to various embodiments of the present disclosure.

[0031] Figure 4a and Figure 4b is an example illustration of a vehicle determining an MRM strategy based on surrounding environment information within a specified MRC range according to various embodiments of the present disclosure.

[0032] Figure 5 is an example illustration of changing the order of priorities of an MRM strategy according to adjacent object information within a specified MRC range in a vehicle according to various embodiments of the present disclosure.

[0033] Figures 6a to 6c is an example illustration of a vehicle determining the possibility of a collision with an adjacent vehicle due to the MRM of the present vehicle according to various embodiments of the present disclosure.

[0034] Figure 7 is an example illustration of determining an MRM strategy of the present vehicle considering the possibility of a collision of the vehicle with an adjacent vehicle due to the MRM of the present vehicle according to various embodiments of the present disclosure.

[0035] Figure 8a and Figure 8b is an example illustration of a vehicle calculating the distance to an adjacent vehicle according to various embodiments of the present disclosure.

[0036] Figure 9 is a flowchart showing the operation of a vehicle according to various embodiments of the present disclosure.

[0037] Figure 10is a flowchart showing determination of an MRM strategy by a vehicle according to various embodiments of the present disclosure.

[0038] Figure 11 is a flowchart showing operations for searching for potential MRC regions and selecting a target MRC region during autonomous driving according to various embodiments of the present disclosure.

[0039] Figure 12 and Figure 13 is a diagram for describing a region of interest (ROI), a potential MRC region (PMZ), and a target MRC region (TMZ) according to various embodiments of the present disclosure.

[0040] Figure 14 is a flowchart showing operations for searching for potential MRC regions and selecting a target MRC region during autonomous driving according to various embodiments of the present disclosure. DETAILED DESCRIPTION

[0041] Hereinafter, embodiments will be described in more detail with reference to the accompanying drawings.

[0042] The configuration and operational effects of the present disclosure will be clearly understood from the following detailed description. Before describing the exemplary embodiments of the present disclosure in detail, it should be noted that throughout the drawings, when possible, the same components will be denoted by the same reference numerals, and detailed descriptions of existing components and functions will be omitted when the subject matter of the present disclosure may be obscured by the description.

[0043] It should also be noted that the terms used in the detailed description of the present disclosure are defined as follows.

[0044] A vehicle refers to a vehicle provided with an autonomous driving system (ADS) and capable of autonomous driving. For example, without the operation of a driver, through the ADS, the vehicle can perform at least one of steering, acceleration, deceleration, lane change, and vehicle stop (short stop). For example, the ADS may include at least one of a pedestrian detection and collision mitigation system (PDCMS), a lane change decision assistance system (LCDAS), a land departure warning system (LDWS), an adaptive cruise control (ACC), a lane keeping assistance system (LKAS), a road boundary departure prevention system (RBDPS), a curve speed warning system (CSWS), a forward vehicle collision warning system (FVCWS), and a low speed following (LSF).

[0045] A driver is a person who uses the vehicle and is provided with the service of the autonomous driving system.

[0046] Vehicle control authority is the authority to control at least one component of a vehicle and / or at least one function of the vehicle. At least one function of the vehicle may include, for example, at least one of steering, acceleration, deceleration (or braking), lane change, lane detection, lateral control, obstacle recognition and distance detection, powertrain control, safety zone detection, engine on / off, power supply on / off, and vehicle locking / unlocking. The listed functions of the vehicle are only examples for helping understanding, and embodiments of the present disclosure are not limited thereto.

[0047] The shoulder may represent the space between the outermost road boundary (or the boundary of the outermost lane) in the traveling direction of the vehicle and the road edge (e.g., curb, guardrail).

[0048] Figure 1 is a block diagram of a vehicle according to various embodiments of the present disclosure. Figure 1 The configuration of the vehicle shown in is an embodiment, and each component may be configured as a chip, a component, or an electronic circuit, or a combination of a chip, a component, and / or an electronic circuit. According to an embodiment, Figure 1 Some of the components shown in may be divided into multiple components and configured as different chips, different components, or different electronic circuits, and some components may be combined to form a chip, a component, or an electronic circuit. According to an embodiment, some of the components shown in may be omitted Figure 1 or other components not shown may be added. Reference will be made to Figure 2 to FIG. 8 to describe Figure 1 at least some of the components. Figure 2 is a functional block diagram of a processor according to various embodiments of the present disclosure, and Figure 3 is a diagram showing a minimum risk maneuver (MRM) strategy according to the vehicle state according to various embodiments of the present disclosure. Figure 4a and Figure 4b are example illustrations of determining an MRM strategy by a vehicle based on surrounding environment information within a specified MRC according to various embodiments of the present disclosure, and Figure 5 is an example illustration of changing the priority order of an MRM strategy according to adjacent object information within a specified MRC in a vehicle according to various embodiments of the present disclosure. Figures 6a to 6c is an example illustration of determining the possibility of a collision with an adjacent vehicle caused by the MRM of the present vehicle by a vehicle according to various embodiments of the present disclosure, and Figure 7 is an example illustration of determining the MRM strategy of the present vehicle by considering the possibility of a collision with an adjacent vehicle caused by the MRM of the present vehicle according to various embodiments of the present disclosure. Figure 8a and Figure 8bThis is an exemplary illustration of calculating the distance to an adjacent vehicle by a vehicle according to various embodiments of the present disclosure.

[0049] Referring to Figure 1 , vehicle 100 may include a sensor unit 110, a controller 120, a processor 130, a display 140, a communication device 150, and a memory 160.

[0050] According to various embodiments, the sensor unit 110 may sense the environment around the vehicle 100 using at least one sensor and generate data related to the environment around the vehicle 100 based on the sensing result. According to an embodiment, the sensor unit 110 may obtain road information, information about objects around the vehicle (e.g., other vehicles, people, objects, curbs, guardrails, lanes, obstacles), and / or the position information of the vehicle based on the sensing data obtained from at least one sensor. For example, the road information may include at least one of lane position, lane shape, lane color, lane type, number of lanes, presence of a shoulder, or size of the shoulder. The objects around the vehicle may include, for example, at least one of the position of the object, the size of the object, the shape of the object, the distance to the object, and the relative speed with respect to the object.

[0051] According to an embodiment, the sensor unit 110 may include, for example, at least one selected from a camera, a light detection and ranging (LIDAR) sensor, a radio detection and ranging (RADAR) sensor, an ultrasonic sensor, an infrared sensor, and a position measurement sensor. The listed sensors are only examples for helping understanding, and the sensors included in the sensor unit 110 of the present disclosure are not limited thereto. The camera may generate image data including objects located in front of, behind, and on the sides of the vehicle 100 by capturing images around the vehicle. The lidar may generate information about objects located in front of, behind, and / or on the sides of the vehicle 100 using light (or laser). The radar may generate information about objects located in front of, behind, and / or on the sides of the vehicle 100 using electromagnetic waves (or radio waves). The ultrasonic sensor may generate information about objects located in front of, behind, and / or on the sides of the vehicle 100 using ultrasonic waves. The infrared sensor may generate information about objects located in front of, behind, and / or on the sides of the vehicle 100 using infrared rays. The position measurement sensor may measure the current position of the vehicle 100. The position measurement sensor may include at least one of a global positioning system (GPS) sensor, a differential global positioning system (DGPS) sensor, and a global navigation satellite system (GNSS) sensor. The position measurement sensor may generate the position data of the vehicle based on signals generated by at least one of the GPS sensor, the DGPS sensor, and the GNSS sensor.

[0052] According to various embodiments, the controller 120 may control the operation of at least one component of the vehicle 100 and / or at least one function of the vehicle according to the control of the processor 130. The at least one function may be, for example, at least one of a steering function, an acceleration function (or a longitudinal acceleration function), a deceleration function (or a longitudinal deceleration function, a braking function), a lane change function, a lane detection function, an obstacle recognition and distance detection function, a lateral control function, a powertrain control function, a safe area detection function, an engine on / off, a power on / off, and a vehicle lock / unlock function.

[0053] According to an embodiment, the controller 120 may control at least one component of the vehicle and / or at least one function for autonomous driving of the vehicle and / or minimum risk maneuver (MRM) of the vehicle 100 according to the control of the processor 130. For example, for minimum risk maneuver, the controller 120 may control the operation of at least one of a steering function, an acceleration function, a deceleration function, a lane change function, a lane detection function, a lateral control function, an obstacle recognition and distance detection function, a powertrain control function, and a safe area detection function.

[0054] According to various embodiments, the processor 130 may control the overall operation of the vehicle 100. According to an embodiment, the processor 130 may include an electronic control unit (ECU) capable of overall controlling components in the vehicle 100. For example, the processor 130 may include a central processing unit (CPU) or a micro processing unit (MCU) capable of performing arithmetic processing.

[0055] According to various embodiments, when a specific event is generated, the processor 130 may activate the autonomous driving system (ADS) to control components in the vehicle 100 so that the vehicle performs autonomous driving. The specified event may be generated when autonomous driving is requested from the driver, vehicle control authority is entrusted from the driver, or conditions specified by the driver and / or the designer are met.

[0056] According to various embodiments, during autonomous driving, the processor 130 may determine whether normal autonomous driving is possible based on at least one of vehicle state information and surrounding environment information. When normal autonomous driving is not feasible, the processor 130 may determine an MRM strategy and control the determined MRM strategy to be executed. Here, the MRM strategy may include an MRM type.

[0057] According to an embodiment, the processor 130 may include a vehicle state information acquisition unit 1310, a surrounding environment information acquisition unit 1320, an MRM strategy determination unit 1330, a potential MRC area search unit 1340, a target MRC area determination unit 1350, and an MRM control unit 1360, as shown in Figure 2 as follows.

[0058] When the ADS is activated, the vehicle state information acquisition unit 1310 may acquire vehicle state information indicating whether there is a mechanical and / or electrical failure in components inside the vehicle by monitoring the mechanical and / or electrical states of components inside the vehicle (e.g., sensors, actuators, etc.). The vehicle state information may include information about the mechanical state and / or electrical state of components inside the vehicle. For example, the vehicle state information may include information indicating whether functions required for autonomous driving can operate normally according to the mechanical and / or electrical states of components inside the vehicle.

[0059] The surrounding environment information acquisition unit 1320 may obtain the surrounding environment information of the vehicle using the sensor unit 110 and / or the communication device 150 from the time when the ADS is activated. The surrounding environment information acquisition unit 1320 may include: a road information acquisition unit 1321 for obtaining information related to the road on which the vehicle is traveling; and an adjacent object information acquisition unit 1322 for detecting objects around the vehicle from the sensor unit 110.

[0060] According to an embodiment, the road information acquisition unit 1321 may acquire the road information of the position where the vehicle is traveling through the sensor unit 110. According to an embodiment, the road information acquisition unit 1321 may acquire map information from an external device (e.g., another vehicle or server) through the communication device 150, and acquire the road information of the position where the vehicle is traveling from the map information.

[0061] According to an embodiment, the adjacent object information acquisition unit 1322 may acquire information about objects around the vehicle (e.g., other vehicles, people, objects, roadside, guardrails, lanes, obstacles) through the sensor unit 110. For example, the adjacent object information acquisition unit 1322 may acquire the distance to and relative speed of at least one vehicle located at the front outer side, side, and / or rear outer side of the vehicle.

[0062] According to an embodiment, the processor 130 may determine whether functions required for autonomous driving can operate normally based on the vehicle state information. The functions required for autonomous driving may include, for example, at least one of a lane detection function, a lane change function, a lateral control function, a deceleration (or braking control) function, a powertrain control function, a safety zone detection function, an obstacle recognition and distance sensing function. When the normal operation of at least one of the functions required for autonomous driving is not feasible, the processor 130 may determine that normal autonomous driving is not feasible.

[0063] According to an embodiment, the processor 130 may determine whether the vehicle state is suitable for general driving conditions based on the vehicle state information. For example, the processor 130 may determine whether the mechanical state information of the vehicle (such as tire pressure information or engine overheat information) is suitable for general driving conditions. When the vehicle state is not suitable for general driving conditions, the processor 130 may determine that normal autonomous driving is not feasible. For example, when the vehicle cannot be driven due to tire pressure or engine overheat, the processor 130 may determine that normal autonomous driving is not feasible.

[0064] According to this embodiment, the processor 130 may determine whether the environment around the vehicle is suitable for the operational design domain (ODD) of autonomous driving based on at least one of the surrounding environment information. The operational design domain may represent the conditions of the surrounding environment for normal operation of autonomous driving. When the surrounding environment information of the vehicle does not match the operational design domain, the processor 130 may determine that normal autonomous driving is not feasible.

[0065] According to various embodiments, when normal autonomous driving is not feasible, the processor 130 may determine it as a situation that requires MRM to minimize the accident risk. In a case where MRM needs to be executed, the processor 130 may select a policy from multiple MRM policies through the use of the MRM policy determination unit 1330. The MRM policy may include three types, as Figure 3 shown. For example, the MRM policy may include a traffic lane parking policy 301 (including type 1 and type 2) and a road shoulder parking policy 303 (including type 3).

[0066] The traffic lane parking policy 301 may include a straight-line parking 311 of type 1 and an in-lane parking 312 of type 2. The road shoulder parking policy 303 may include a half-road shoulder parking 313 of type 3 and a full-road shoulder parking 314 of type 4.

[0067] The straight-line parking 311 of type 1 is a type that stops the vehicle using a deceleration control 323, which is only a longitudinal deceleration function and is not accompanied by lateral control. For example, when the lateral control 321, acceleration control 322, lane change 324, and detection of potential parking positions 325 outside the traffic lane are not feasible and only deceleration (or longitudinal deceleration) is feasible, the type of straight-line parking 311 can be executed. For example, in a case where the lane cannot be detected due to a defect in the actuator and lateral control cannot be performed, the straight-line parking type can be executed. Here, the detection of potential parking positions outside the traffic lane may be a function for detecting a safe area (such as the road shoulder) or a rest area located outside the traffic lane.

[0068] Lane parking 312 of type 2 is a type in which a vehicle parks within the boundaries of the lane on which it is traveling. For example, lane parking 312 may refer to a type in which a vehicle stops within the boundaries of the lane on which it is traveling through lateral control 321 and / or deceleration control 323. The driving lane may refer to the lane on which the vehicle is traveling when it is determined that MRM is required. Lane parking 312 may be performed when at least one of the functions among acceleration control 322, lane change 324, or detection of a potential parking position 325 outside the traffic lane is not feasible.

[0069] Shoulder parking 313 of type 3 is a type in which a vehicle is stopped in a state where a part of the vehicle is located on the shoulder of the road. For example, shoulder parking 313 may be a type in which, after a part of the vehicle moves to the shoulder of the road (or the boundary of the outermost lane) outside the boundary of the road through lateral control 321, deceleration control 323, lane change 324, and / or detection of a potential stop position 325 outside the traffic lane, the vehicle is stopped.

[0070] Full shoulder parking 314 of type 3 is a type in which a vehicle is stopped in a state where the entire vehicle is located on the shoulder of the road. For example, full shoulder parking 314 may be a type in which the vehicle is stopped through lateral control 321, deceleration control 323, lane change 324, and / or detection of a potential vehicle position 325 outside the traffic lane, such that the entire vehicle moves to the shoulder of the road to be positioned on the shoulder of the road outside the boundary of the road.

[0071] According to an embodiment, the priority of the above MRM types may be determined based on the road, the surrounding environment, and the fail-operational ability as a limit of vehicle tolerance for faults. For example, the priority of the MRM type corresponding to the shoulder parking strategy 303 may be set higher than the priority of the MRM type corresponding to the traffic lane parking strategy 301 in order to minimize danger when stopping the vehicle. In addition, the priority of full shoulder parking 314 may be set higher than the priority of shoulder parking 313, and the priority of lane parking 312 may be set higher than the priority of straight parking 311. That is, the priority of the MRM types may be set to decrease in the order of full shoulder parking 314, shoulder parking 313, lane parking 312, and straight parking 311.

[0072] Referring again to Figure 2 , according to various embodiments, the MRM strategy determination unit 1330 may select an MRM strategy based on at least one of vehicle state information and surrounding environment information.

[0073] According to an embodiment, the MRM policy determination unit 1330 may check the MRM types that can be executed among the above MRM types based on the vehicle state information, by examining the functions that operate normally and / or the functions that do not operate normally among the functions required for autonomous driving. For example, when the lateral control function operates normally, it may be determined that all MRM types can be executed, that is, straight parking 311, in-lane parking 312, partial shoulder parking 313, and full shoulder parking 314. As another example, when the lateral control function does not operate normally, straight parking 311 may be determined as the MRM type that can be executed.

[0074] If there is one MRM type that can be executed based on the vehicle state information, the MRM policy determination unit 1330 may determine the corresponding MRM type as the MRM policy. For example, in the case where the lateral control function does not operate normally, only straight parking 311 can be executed, and thus straight parking 311 may be determined as the MRM policy. As another example, when the driving lane is not detected due to sensor defects and / or external environment, the MRM policy determination unit 1330 may only execute straight parking 311, and thus, straight parking 311 may be determined as the MRM policy.

[0075] When there are multiple MRM types that can be executed based on the vehicle state information, the MRM policy determination unit 1330 may determine the MRM types that can be executed within the specified minimum risk condition (MRC) range. According to an embodiment, the specified MRC range may be set and / or changed by the operator and / or the designer. According to an embodiment, the specified MRC range may be set differently according to vehicle performance, vehicle type, and / or external environment factors (e.g., weather, time, etc.).

[0076] According to an embodiment, the MRM policy determination unit 1330 may determine the MRM types that can be executed within the MRC range based on whether there is a shoulder within the specified MRC range. When there is no shoulder within the specified MRC range, the MRM policy determination unit 1330 may determine in-lane parking 312 or straight parking 311 as the MRM types that can be executed within the MRC range.

[0077] When there is a shoulder within the specified MRC range, the MRM policy determination unit 1330 can determine the type of MRM that can be executed within the MRC range based on the size of the shoulder. When the size of the shoulder within the specified MRC range is greater than or equal to the specified size, the MRM policy determination unit 1330 can determine full-shoulder parking 314, half-shoulder parking 313, in-lane parking 312, or straight-line parking 311 as the type of MRM that can be executed within the MRC range. The specified size can be determined based on the size of the vehicle. When the size of the shoulder is less than the specified size, the MRM policy determination unit 1330 can determine half-shoulder parking 313, in-lane parking 312, or straight-line parking 311 as the type of MRM that can be executed within the MRC range.

[0078] According to various embodiments, when there are multiple types of MRMs that can be executed within the specified MRC range, the MRM policy determination unit 1330 can consider priority and / or adjacent object information to select the final MRM policy.

[0079] According to an embodiment, when there are multiple types of MRMs that can be executed within the specified MRC range, the MRM policy determination unit 1330 can determine the MRM type with the highest priority among the types of MRMs that can be executed within the specified MRC range as the final MRM policy. For example, as Figure 4a shown, when the width of the shoulder 410 within the specified MRC range 400 is greater than the width of the vehicle 100, the MRM policy determination unit 1330 can determine full-shoulder parking 314, which has the highest priority among the types of MRMs that can be executed within the MRC range 400, as the final MRM policy. As another example, as Figure 4b shown, when the width of the shoulder 420 within the specified MRC range 400 is less than the width of the vehicle 100, the MRM policy determination unit 1330 can determine half-shoulder parking 313, which has the highest priority among the types of MRMs that can be executed within the specified MRC range 400, as the final MRM policy.

[0080] According to an embodiment, the MRM policy determination unit 1330 can determine the final MRM policy by additionally considering the risk associated with executing the MRM policy within the specified MRC range. For example, as Figure 5As shown, it is assumed that the road shoulder 501 exists within the MRC range 400. However, among the areas of the road shoulder 501 within the MRC range 400, the width of the area adjacent to the vehicle 100 is greater than the width of the vehicle 100, and the width of the area located away from the vehicle 100 is less than the width of the vehicle 100. That is, a description is made assuming a situation where a road shoulder 501 with a gradually decreasing width exists within the MRC range 400. In this case, the MRM strategy determination unit 1330 can select the full road shoulder parking 314 with the highest priority based on the width of the road shoulder 501. However, when executing the full road shoulder parking 314, when there is a risk of collision (or impact) 520 with the rear end of another vehicle 510, since the half road shoulder parking 313 does not have a risk of collision with the rear end of another vehicle 510, the MRM strategy determination unit 1330 can select the half road shoulder parking 313 with a lower priority than the full road shoulder parking 314 as the final MRM strategy.

[0081] According to an embodiment, when there are multiple MRM types that can be executed within a specified MRC range, the MRM strategy determination unit 1330 can consider the possibility of collision and / or the existence of accident liability to select the final MRM strategy. The MRM strategy determination unit 1330 can determine the possibility of collision with adjacent vehicles and the existence of accident liability in a collision event based on the driving paths of each MRM type that can be executed within the specified MRC range. Regarding the full road shoulder parking and / or half road shoulder parking that require lane changes, the MRM strategy determination unit 1330 can determine the possibility of collision and / or the liability for an accident with the vehicle located in the front outer side, the side, and / or the rear outer side of the vehicle among adjacent vehicles. Regarding the in-lane parking that does not require lane changes, the MRM strategy determination unit 1330 can determine the possibility of collision and / or the liability for an accident with the vehicle located behind among adjacent vehicles.

[0082] To determine the possibility of collision with adjacent vehicles and / or the existence of accident liability, the MRM strategy determination unit 1330 can calculate a safety distance representing the difference between the minimum relative distance and the actual relative distance of adjacent vehicles based on the Responsibility-Sensitive Safety (RSS) model shown in Mathematical Equations 1 and 2, and can determine the possibility of collision and the existence of accident liability based on the calculated safety distance.

[0083]

[0084]

[0085] ……Equation (1)

[0086] Here, RSS x represents the longitudinal safety distance, dmin, x represents the minimum longitudinal relative distance to an adjacent vehicle to be maintained, d x represents the actual longitudinal relative distance between the host vehicle and the adjacent vehicle. Additionally, RSS y represents the lateral safety distance, d min, y represents the minimum lateral relative distance to an adjacent vehicle to be maintained, d y represents the actual lateral relative distance between the host vehicle and the adjacent vehicle.

[0087] When at least one of the longitudinal safety distance (RSS x ) and the lateral safety distance (RSS y ) to an adjacent vehicle is positive, even if an MRM for a driving path related to the adjacent vehicle is executed, the MRM policy determination unit 1330 can determine that the possibility of a collision with the adjacent vehicle is low (or no collision is possible). Additionally, even if a collision occurs with the adjacent vehicle, the MRM policy determination unit 1330 can determine that the vehicle is not responsible for the accident. For example, as Figure 6a shown, if the longitudinal safety distance (RSS x ) from vehicle 100 to the vehicle 601 in the right front is negative, but its lateral safety distance (RSS y ) is positive, even if an MRM that requires a lane change (e.g., half-shoulder stop, and / or full-shoulder stop) is executed, it can be determined that the possibility of a collision with the vehicle 601 in the right front is low, and also, even if the vehicle collides with the vehicle 601 in the right front, it can be determined that the host vehicle 100 is not responsible for the accident. As another example, as Figure 6b shown, if the lateral safety distance (RSS y ) to the vehicle 611 in front traveling in the same lane as the host vehicle 100 is negative, but its longitudinal safety distance (RSS x ) is positive, even if an MRM that requires a lane change (e.g., half-shoulder stop, and / or full-shoulder stop) is executed, it can be determined that the possibility of a collision with the vehicle 611 in front is low, and even if the vehicle collides with the vehicle 611 in front, it can be determined that the host vehicle 100 is not responsible for the accident.

[0088] When both the longitudinal safety distance (RSS x ) and the lateral safety distance (RSS y ) to an adjacent vehicle are negative and an MRM for a driving path related to the adjacent vehicle is executed, the MRM policy determination unit 1330 can determine that the possibility of a collision with the adjacent vehicle is high (or there is a possibility of a collision). Additionally, when the vehicle collides with the adjacent vehicle, the MRM policy determination unit 1330 can determine that the host vehicle is responsible for the accident. For example, as Figure 6c shown, when the longitudinal safety distance (RSSx ), and the lateral safety distance (RSS y ), when both are negative, when performing an MRM that requires a lane change (e.g., a half-shoulder stop and / or a full-shoulder stop), the possibility of collision with the vehicle 621 in the right front can be determined to be high. In addition, since the lane change of the present vehicle can be the cause of the collision with the vehicle 621 in the right front, when a collision with the vehicle 621 in the right front occurs, the MRM strategy determination unit 1330 can determine that the accident liability lies with the present vehicle.

[0089] When performing an MRM that requires a lane change, when it is determined that the possibility of collision with an adjacent vehicle is high, the MRM strategy determination unit 1330 can select a stop within the lane with a lower priority than a full-shoulder stop and a half-shoulder stop as the final MRM. In this case, the MRM strategy determination unit 1330 can calculate the longitudinal safety distance and the lateral safety distance from the vehicle behind traveling in the same lane as the present vehicle. The MRM strategy determination unit 1330 can determine that when at least one of the longitudinal safety distance and the lateral safety distance from the vehicle behind is positive, even if a stop within the lane is performed, the possibility of collision with the vehicle behind is low, and even if the vehicle collides with it, the accident liability does not lie with the present vehicle, and thus, the MRM strategy determination unit 1330 can select a stop within the lane as the final MRM. For example, as Figure 7 shown, the present vehicle 100 can calculate the longitudinal safety distance and the lateral safety distance from the rear outer vehicle 720 in a case where an MRM needs to be performed to perform a type 3 full-shoulder stop with the highest priority. However, when both the longitudinal safety distance and the lateral safety distance from the rear outer vehicle 720 are negative, due to the lane change required when performing a full-shoulder stop, there is a high possibility of collision with the rear outer vehicle 720, and in the case of a collision with the rear outer vehicle 720, the present vehicle may have the liability for the collision. Therefore, since both the longitudinal safety distance and the lateral safety distance between the present vehicle 100 and the vehicle 710 behind are positive, the present vehicle 100 can select a type 2 stop within the lane with a lower priority than type 4 as the final MRM strategy.

[0090] According to an embodiment, the longitudinal safety distance and the lateral safety distance between the present vehicle and an adjacent vehicle can be calculated by the following mathematical equations 2 and 3.

[0091] The following mathematical equation 2 is as Figure 8a shown for calculating the longitudinal safety distance (Rss x ) 810 between the present vehicle Cr and the adjacent vehicle Cf, and the mathematical equation 3 is as Figure 8b shown for calculating the lateral safety distance (Rss y ) 820 between the present vehicle Cr and the adjacent vehicle Cf.

[0092]

[0093] ……Equation (2)

[0094]

[0095] ……Equation (3)

[0096] Here, ρ can represent the reaction time, μ can represent the lateral margin, a min，brake can represent the minimum deceleration of this vehicle, a max，aceel can represent the maximum acceleration of the adjacent vehicle, and a max，brake can represent the maximum deceleration of the adjacent vehicle.

[0097] To calculate the lateral safety distance and / or the longitudinal safety distance, the parameters of mathematical equations 2 and 3 can be set as shown in Table 1 below.

[0098] Table 1

[0099]

[0100] The parameter values in Table 1 are not limited to this.

[0101] According to the embodiment, when selecting the final MRM strategy, the MRM strategy determination unit 1330 can store the basic data for selecting the final MRM strategy in the memory 160. The basic data can include at least one of the presence of the road shoulder within the specified MRC range, the size of the road shoulder (e.g., length and / or width), the safety distance from the adjacent vehicle, the vehicle state information of this vehicle, and the lane detection information. The MRM strategy determination unit 1330 can ensure the basis for selecting the MRM strategy with a lower priority by storing the basic data for which it selects the final MRM strategy. For example, when selecting the straight parking of type 1 as the final MRM strategy, the MRM strategy determination unit 1330 can store in the memory 160 at least one of the information indicating the abnormal operation of the lateral control function, the steering angle, the steering speed, the information indicating the defect of the lane detection sensor, and the sensed value of the lane detection sensor. As another example, even if there is a road shoulder within the specified MRC range, when selecting the in-lane parking of type 2 as the final MRM strategy, the MRM strategy determination unit 1330 can also store the information about the longitudinal safety distance and the lateral safety distance to at least one adjacent vehicle calculated in the MRM strategy selection process.

[0102] If there is an MRM type that can be executed based on vehicle state information, the MRM policy determination unit 1330 may determine the corresponding MRM type as the MRM policy. For example, the MRM policy determination unit 1330 can only execute the straight-line parking 311 when the lateral control function cannot operate normally, so the straight-line parking 311 can be determined as the MRM policy. As another example, when the driving lane is not detected due to sensor defects and / or the external environment, the MRM policy determination unit 1330 may determine the straight-line parking 311 as the MRM policy.

[0103] According to various embodiments, the processor 130 may control to stop the vehicle according to the final MRM policy, and at the same time control to notify other vehicles and / or drivers of the information indicating that the MRM is in progress. The control operation to stop the vehicle may include generating a driving trajectory for stopping the vehicle and / or lateral and / or longitudinal control following the generated driving trajectory. The processor 130 may control the display 140 to notify the driver that the vehicle is performing the MRM. As another example, the processor 130 may control the communication device 150 to notify other vehicles that the vehicle is performing the MRM. This is only an example for helping understanding, and the method for notifying that the MRM is in progress will not be limited thereto.

[0104] According to various embodiments, the processor 130 may perform a control operation for stopping the vehicle according to the determined MRM type, and determine whether the MRC is satisfied. The MRC may represent a parking state where the vehicle speed is 0. For example, the processor 130 may determine whether the vehicle 100 enters a parking state where the speed of the vehicle 100 is 0 while performing at least one operation according to the determined final MRM type. When the vehicle 100 enters a state where the speed is 0, the processor 130 may determine that the MRC is satisfied.

[0105] According to various embodiments, when the MRC is satisfied, the processor 130 may end the MRM operation and switch the automatic driving system (ADS) to the standby mode or the off state. According to an embodiment, after switching the autonomous driving system (ADS) to the standby mode or the off state, the processor 130 may control the vehicle control authority to transfer to the driver (or user).

[0106] According to various embodiments, when the determined MRM type is the roadside parking type, the processor 130 may set an area of interest (ROI) based on pre-input information. The detailed description thereof is provided below.

[0107] According to an embodiment, the potential MRC area search unit 1340 may search for a potential MRC area (PMZ) that is a candidate area for shoulder parking within the ROI. According to an embodiment, the target MRC area determination unit 1350 may select a target MRC area (TMZ) from the potential MRC area (PMZ). A detailed description thereof is provided below.

[0108] According to an embodiment, when determining the target MRC area (PMZ), the MRM control unit 1360 may control the vehicle to move into the target MCR area (TMZ). A detailed description thereof is provided below.

[0109] Reference Figure 1 , the display 140 visually displays information related to the vehicle 100. For example, under the control of the processor 130, the display 140 may provide various information related to the state of the vehicle 100 to the driver of the vehicle 100. The various information related to the state of the vehicle may include at least one of information indicating whether various components included in the vehicle and / or at least one function of the vehicle are operating normally, and information indicating the driving state of the vehicle. The driving state of the vehicle may include, for example, at least one of the state of the vehicle's autonomous driving, the state in which the MRM is in progress, the state in which the MRM is completed, and the state in which the autonomous driving ends.

[0110] According to various embodiments, the communication device 150 may communicate with an external device of the vehicle 100. According to an embodiment, under the control of the processor 130, the communication device 150 may receive data from an external device of the vehicle 100 or transmit data to an external device of the vehicle 100. For example, the communication device 150 may perform communication using a wireless communication protocol or a wired communication protocol.

[0111] In the above Figure 1 , the controller 120 and the processor 130 have been described as separate components, but according to various embodiments, the controller 120 and the processor 130 may be integrated into one component.

[0112] Figure 9 is a flowchart showing the operation of a vehicle according to various embodiments of the present disclosure. Figure 9 The vehicle of Figure 1 may be the vehicle 100 of

[0113] Reference Figure 9 , the vehicle 100 may normally operate the ADS in operation S910.

[0114] According to an embodiment, the vehicle 100 may monitor the vehicle state and the surrounding environment while performing autonomous driving according to the normal operation of the ADS. The vehicle 100 may sense whether MRM is required based on the information obtained by monitoring the vehicle state and the surrounding environment. If MRM is required, an event A1 may be generated.

[0115] According to an embodiment, the vehicle 100 may sense whether driver (or user) intervention is required when performing autonomous driving according to the normal operation of the ADS. When driver intervention is required, the vehicle 100 may perform a request to intervene (RTI) or issue a warning through the ADS. The driver intervention request or warning may be an event A2. When event A1 is generated in the state of normal operation of the ADS, the vehicle 100 may proceed to operation S920.

[0116] When event A2 is generated in the state of normal operation of the ADS, in operation S950, the vehicle 100 may determine whether driver intervention is sensed within a specified time. When driver intervention is not sensed within the specified time, the vehicle 100 may determine that event B1 has occurred. When event B1 occurs, the vehicle 100 may proceed to operation S920. When driver intervention is sensed within the specified time, the vehicle 100 may determine that event B2 has occurred. When event B2 occurs, the vehicle 100 may proceed to operation S940.

[0117] In operation S920, the vehicle 100 may perform MRM. According to an embodiment, the vehicle 100 may determine the MRM type based on at least one of the vehicle state information and the surrounding environment information. The surrounding environment information may include information about the road and information about adjacent vehicles. As Figure 3 shown, the MRM type may include straight parking 311 of type 1, in-lane parking 312 of type 2, half-road shoulder parking 313 of type 3, and full-road shoulder parking 314 of type 3. The vehicle 100 may control at least one component in the vehicle to stop the vehicle according to the determined MRM type. According to an embodiment, the vehicle 100 may store basic data for determining the final MRM type in the memory 160.

[0118] In operation S920, when the speed of the vehicle becomes 0 by performing MRM, the vehicle 100 may determine whether the minimum risk condition is satisfied. When the minimum risk condition is satisfied, the vehicle 100 may determine that event C1 has occurred and may proceed to operation S930. The vehicle 100 may determine whether driver intervention is sensed during the MRM. When driver intervention is sensed, the vehicle 100 may determine that event C2 has occurred and may proceed to operation S940.

[0119] In operation S930, vehicle 100 may maintain a state that satisfies the minimum risk condition. A state that satisfies the minimum risk condition may represent a state where the vehicle is parked. For example, vehicle 100 may maintain a parked state. For example, vehicle 100 may perform a control operation to keep the vehicle in a parked state regardless of the inclination of the road surface at the parking position. Vehicle 100 may determine whether event D1 occurs while maintaining a state that satisfies the minimum risk condition. Event D1 may include at least one of the ADS being turned off by the driver and the completion of the transfer of vehicle control to the driver. When event D1 occurs, vehicle 100 may proceed to operation S940.

[0120] In operation S940, vehicle 100 may switch the ADS to the standby mode or the off state. When the ADS is in the standby mode or the disconnected state, vehicle 100 may not perform operations for autonomous driving.

[0121] The above operations S910, S920, S930, and S950 may be in a state where the ADS is activated, and operation S940 may be in a state where the ADS is deactivated.

[0122] Figure 10 is a flowchart showing the determination of the MRM strategy by a vehicle according to various embodiments of the present disclosure. Figure 10 The operation of may be Figure 9 The detailed operation of operation S920 of. In the following embodiments, each operation may be executed sequentially, but does not have to be executed sequentially. For example, the order of each operation may be changed, and at least two operations may be executed in parallel. In addition, the operations described below may be executed by the processor 130 and / or the controller 120 provided in the vehicle 100 or implemented as instructions executable by the processor 130 and / or the controller 120.

[0123] Refer to Figure 10 , in operation S1001, vehicle 100 may determine whether lateral control is feasible based on vehicle state information. For example, vehicle 100 may obtain vehicle state information indicating whether a mechanical and / or electrical failure of vehicle internal components (such as sensors, actuators, etc.) occurs by monitoring the mechanical and / or electrical states of vehicle internal components. Vehicle 100 may determine whether the lateral control (or steering control) of vehicle 100 is feasible based on the vehicle state information indicating the mechanical state and / or electrical state of the sensor and / or actuator.

[0124] If the lateral control is not feasible, then in operation S1021, vehicle 100 may select straight-line parking as the final MRM strategy. For example, as Figure 3 shown, since vehicle 100 that cannot perform lateral control may only perform straight-line parking, vehicle 100 may determine straight-line parking as the final MRM strategy.

[0125] If lateral control is feasible, in operation S1003, vehicle 100 may determine whether the road shoulder of the road exists within the MRC range. For example, vehicle 100 may determine whether there is a road shoulder by checking the road information within the MRC range, and the MRC range corresponds to within a specified distance centered on the vehicle 100 itself. The road information within the MRC range may be obtained from the sensing data of sensors (such as sensor unit 110) provided in vehicle 100, or from the map information obtained through communication device 150.

[0126] When there is no road shoulder, in operation S1015, vehicle 100 may determine whether lane detection is feasible. For example, since vehicle 100 cannot execute the road shoulder parking strategy when there is no road shoulder of the road, vehicle 100 may check whether lane detection is feasible to determine whether parking within the lane is feasible. Vehicle 100 may determine that lane detection is feasible or lane detection is not feasible based on the sensed value of the lane detection sensor.

[0127] When lane detection is not feasible, in operation S1021, vehicle 100 may select straight parking as the final MRM strategy. For example, when the lane cannot be detected, vehicle 100 may determine that parking within the lane cannot be executed, and may determine straight parking as the final MRM strategy.

[0128] If lane detection is feasible, in operation 1017, vehicle 100 may determine whether the accident liability lies with the vehicle itself during the deployment (or implementation) of parking within the lane. For example, vehicle 100 may calculate the safety distance from the vehicle behind, and determine the possibility of collision with the vehicle behind and whether it is responsible for the accident based on the calculated safety distance. The vehicle behind may represent a vehicle traveling in the same lane as the vehicle itself. The safety distance from the vehicle behind may include a longitudinal safety distance and a lateral safety distance as shown in Mathematical Equation 1. If both the calculated longitudinal safety distance and lateral safety distance are negative, when parking within the lane is executed, vehicle 100 is very likely to collide with the vehicle behind, and when a collision with the vehicle behind occurs, it can be determined that the accident liability lies with the vehicle itself. When at least one of the calculated longitudinal safety distance and lateral safety distance is positive, vehicle 100 is unlikely to collide with the vehicle behind during the implementation of parking within the lane, and when a collision with the vehicle behind occurs, it can be determined that the vehicle has no accident liability.

[0129] When it is determined that the accident liability lies with the vehicle itself during the completion of parking within the lane, vehicle 100 may proceed to operation S1021 and select straight parking as the final MRM strategy.

[0130] When it is determined that the accident liability does not lie with the vehicle during parking within the lane, in operation S1019, vehicle 100 may select parking within the lane as the final MRM strategy.

[0131] If, as a result of the check in operation S1003, there is a road shoulder, in operation S1005, vehicle 100 may determine whether the size of the road shoulder is greater than the size of the vehicle. For example, vehicle 100 may compare the width of the road shoulder with the width of the vehicle to determine whether the vehicle can perform a full-road-shoulder parking or a half-road-shoulder parking.

[0132] When the size of the road shoulder is greater than the size of the vehicle, vehicle 100 may determine that full-road-shoulder parking is feasible, and when performing full-road-shoulder parking in operation 1007, the vehicle may determine whether the accident liability lies with the vehicle. For example, vehicle 100 may calculate the driving path for achieving full-road-shoulder parking and may calculate the safety distance from at least one adjacent vehicle related to the calculated driving path. At least one adjacent vehicle related to the driving path for full-road-shoulder parking implementation may include at least one of a vehicle located in the front outer side, a side vehicle, and / or a vehicle located in the rear outer side. Vehicle 100 may determine the likelihood of collision with at least one adjacent vehicle and whether it is responsible for the accident based on the calculated safety distance. The safety distance from at least one adjacent vehicle may include a longitudinal safety distance and a lateral safety distance, as shown in mathematical equation 1. If both the calculated longitudinal safety distance and lateral safety distance are negative, when performing full-road-shoulder parking, vehicle 100 is very likely to collide with at least one adjacent vehicle, and when a collision with the vehicle occurs, it may be determined that the accident liability lies with the vehicle. If at least one of the calculated longitudinal safety distance and lateral safety distance is positive, vehicle 100 may determine that the likelihood of collision between vehicle 100 and at least one adjacent vehicle during full-road-shoulder parking is low, and if a collision with the vehicle occurs, the accident liability does not lie with the vehicle.

[0133] When it is determined that the accident liability does not lie with the vehicle during completion of full-road-shoulder parking, vehicle 100 may select full-road-shoulder parking as the final MRM strategy in operation S1009.

[0134] When it is determined that the accident liability lies with the vehicle itself during full shoulder parking, vehicle 100 may proceed to operation S1011 to determine whether the accident liability lies with the vehicle itself during half shoulder parking. For example, vehicle 100 may calculate the driving path for achieving half shoulder parking and may calculate the safety distance from at least one adjacent vehicle related to the calculated driving path. At least one adjacent vehicle related to the driving path for achieving half shoulder parking may include at least one of a vehicle located at the front outer side, a side vehicle, and / or a vehicle located at the rear outer side. Vehicle 100 may determine the possibility of collision with at least one adjacent vehicle and whether it is responsible for the accident based on the calculated safety distance. The safety distance from at least one adjacent vehicle may include a longitudinal safety distance and a lateral safety distance, as shown in mathematical equation 1. If both the calculated longitudinal safety distance and lateral safety distance are negative, then when performing half shoulder parking, vehicle 100 is very likely to collide with at least one adjacent vehicle, and when a collision occurs with a vehicle, it may be determined that the accident liability lies with the vehicle itself. If at least one of the calculated longitudinal safety distance and lateral safety distance is positive, vehicle 100 may determine that the possibility of collision between vehicle 100 and at least one adjacent vehicle during the completion of half shoulder parking is low, and if a collision occurs with a vehicle, the accident liability does not lie with the vehicle itself.

[0135] When it is determined that the accident liability does not lie with the vehicle itself during the completion of half shoulder parking, vehicle 100 may select half shoulder parking as the final MRM strategy in operation S1013.

[0136] When it is determined that the accident liability lies with the vehicle itself during half shoulder parking, vehicle 100 may proceed to operation S1017.

[0137] As described above, when a vehicle according to various embodiments of the present disclosure detects a situation where normal autonomous driving is not possible during autonomous driving, the vehicle may determine an MRM strategy based on vehicle state information and / or surrounding environment information, considering the accident liability for each MRM type, thereby improving safety while minimizing the risk of the vehicle.

[0138] Figure 11 is a flowchart showing operations for searching for a potential MRC area and selecting a target MRC area during autonomous driving according to various embodiments of the present disclosure. Figure 12 and Figure 13 is a diagram for describing a region of interest (ROI), a potential MRC area (PMZ), and a target MRC area (TMZ) according to various embodiments of the present disclosure. Figures 11 to 13 The vehicle described in Figure 1 may be vehicle 100.

[0139] Refer to Figure 11, in operation S1110, the processor 130 may determine whether MRM is required based on at least one of the surrounding environment information and the vehicle state information during the autonomous driving of the vehicle 100, and determine the MRM type when it is determined that MRM is required.

[0140] In operation S1120, when the MRM type is the road shoulder parking type, the processor 130 may set a region of interest (ROI) based on pre-input information.

[0141] The road shoulder parking type is an example, and the MRM type may be the half road shoulder parking 313 or the full road shoulder parking 314 type of type 3.

[0142] Reference Figure 12 , as an example, the region of interest (ROI) may have a rectangular shape including an upper side R1, a lower side R2, a left side R3, and a right side R4. The upper side R1 may have a length from a point on the front part of the vehicle 100 (e.g., the hood, the front bumper, the headlights, etc.) to a position spaced apart by a predetermined first distance. The lower side R2 may have the same length as the upper side R1. The left side R3 may have a length from a point on the center line to the boundary of the road shoulder, and the right side R4 may have the same length as the left side R3. As an embodiment, a point on the edge of the front part of the vehicle 100 may be the starting point of the upper side R1, and a point obtained by adding the predetermined first distance to the starting point may be the ending point. As an embodiment, a point on the center line may be the starting point of the left side R3, and a point on the outer boundary of the line of the road shoulder may be the ending point of the left side R3. As an embodiment, the left side R3 may have a length from a point on a line near the outermost line (e.g., the center line) in the traffic lane in which the vehicle 100 travels to the line of the road shoulder that is far from the outermost line (e.g., the center line) (or the outer boundary of the line of the road shoulder). As an embodiment, a point on the outermost line (e.g., the center line) of the traffic lane in which the vehicle 100 travels may be the starting point of the left side R3, and a point on the outer boundary of the line of the road shoulder may be the ending point of the left side R3. As an embodiment, the ROI may have a rectangular shape, but is not limited thereto.

[0143] As an embodiment, the ROI may be a region for performing a search for a potential MRC region (PMZ), and may be a region that does not change even when there are changes in the position according to the movement of the vehicle, changes in the state of the vehicle, and changes in the environment. That is, the ROI may be fixed as the initially set region, even if the vehicle 100 continues to move, and the search for the potential MRC region (PMZ) and the selection of the target MRC region (TMZ) may be performed only within the fixed ROI. Based on the ROI, the generation of the potential MRC region (PMZ) and the target MRC region (TMZ) and the failure response may be performed.

[0144] In operation S1130, the processor 130 may search for potential MRC regions (PMZs) that are candidate MRC regions within the ROI. The processor 130 may use a pre-stored algorithm to search for potential MRC regions (PMZs) based on pre-stored map information, surrounding environment information, and traffic information.

[0145] Reference Figure 12 and Figure 13 , as an example, the processor 130 may determine (or search) within the region of interest (ROI) a region that the vehicle can stop at by reducing speed and moving to a different lane from the vehicle's current position, is not occupied by an obstacle, and is larger than a predetermined reference size as a potential MRC region (PMZ). As an example, the processor 130 may determine whether a designated region within the region of interest (ROI) corresponds to a position where the vehicle can stop by decelerating and moving to a lane different from the vehicle's current position, as a process for searching for potential MRC regions (PMZs), based on pre-stored map information. For example, when the position of a designated region within the region of interest (ROI) based on pre-stored map information is a shoulder position, the processor 130 may determine whether the obstacle occupies the designated region and whether the size of the designated region is larger than a predetermined reference size.

[0146] As an example, the processor 130 may determine whether an obstacle (e.g., another vehicle, person, animal, object, etc.) occupies a designated region within the region of interest (ROI), as a process for searching for potential MRC regions (PMZs), based on surrounding environment information and / or traffic information. As an example, if a designated region within the region of interest (ROI) is not occupied by an obstacle, the processor 130 may determine whether the designated region within the region of interest (ROI) corresponds to a position where the vehicle can stop by decelerating and moving to a lane different from the vehicle's current position, and whether the size of the designated region is larger than a predetermined reference size.

[0147] As an example, when the size of the designated region (where the vehicle can stop by decelerating and moving to a different lane from the vehicle's current position and is not occupied by an obstacle) is larger than a predetermined reference size, the processor 130 may determine the designated region as a potential MRC region (PMZ).

[0148] The potential MRC region (PMZ) does not mean a point, but rather an idle space where the vehicle may park in the shoulder area within the ROI, and may mean the area where the vehicle decelerates to move into the shoulder to park in the shoulder, or the area where the vehicle's posture (e.g., parallel parking) is aligned. As an example, the processor 130 may search for single or multiple potential MRC regions (PMZs).

[0149] As an example, the processor 130 may search for a potential MRC zone (PMZ) for a predetermined time (e.g., 1 second). The situation that triggers the MRM corresponds to a dangerous situation where autonomous driving cannot continue. Therefore, the search for the potential MRC zone (PMZ) must be completed within a short period of time.

[0150] As an example, the processor 130 may continue to search for the potential MRC zone (PMZ) as a background task even after the search for the potential MRC zone (PMZ) is completed. By doing so, when the selection of the target MRC zone (PMZ) fails, a re-search for the potential MRC zone (PMZ) can be quickly performed.

[0151] The potential MRC zone (PMZ) may have multiple candidate MRC zones, and each candidate MRC zone may have a different size.

[0152] As an example, the processor 130 may set the size of the potential MRC zone (PMZ) taking into account the traffic information of the road shoulder or traffic lane.

[0153] As an example, when it is determined that a lane change is not easy due to adjacent vehicles, the processor 130 may search for a potential MRC zone (PMZ) that is spaced a certain distance from the current position of the vehicle 100.

[0154] A clearance area may be formed between the candidate MRC zone near the obstacle and the obstacle among the multiple candidate MRC zones. The potential MRC zone (PMZ) is an area not occupied by obstacles, and a safe clearance area for the obstacles can be formed therein.

[0155] Referring to FIG. 13, based on the obstacle 20 positioned on the road shoulder, the potential MRC zone (PMZ) may include a first candidate MRC zone (PMZ1) in the direction close to the vehicle 100 positioned in front of the obstacle 20 and a second candidate MRC zone (PMZ2) in the direction away from the vehicle 100 positioned behind the obstacle 20. A first clearance area (G1) may be formed between the first candidate MRC zone (PMZ1) and the obstacle 20, and a second clearance area (G2) may be formed between the obstacle 20 and the second candidate MRC zone (PMZ2).

[0156] As an example, the second clearance area (G2) may be larger than the first clearance area (G1). In the case of the first clearance area (G1), the vehicle 100 that has moved onto the shoulder approaches the obstacle 20, and thus a larger clearance area is not required. However, in the case of the second clearance area (G2) behind the obstacle 20, the vehicle 100 moves into the lane of the shoulder while avoiding the obstacle 20, and in this case, the risk of collision is high. Therefore, a relatively larger clearance area than the first clearance area (G1) may be required.

[0157] In operation S1140, the processor 130 may select a target MRC area (TMZ) in the potential MRC area (PMZ). Using a pre-stored algorithm, the processor 130 may be based on information about the size of the potential MRC area (PMZ) (e.g., PMZ) (information about the longitudinal and lateral lengths), information about the distance to the potential MRC area (PMZ), and information about the movement path to the potential MRC area (PMZ), information about the state of the vehicle 100 (e.g., information about the current state of the vehicle (e.g., the severity of the defect) and information about the maximum distance the vehicle may move considering the future state of the vehicle), information about the number of lane changes required to reach the potential MRC area (PMZ), surrounding environment information (e.g., information about the terrain around the potential MRC area (PMZ)), information that affects the safety during parking (such as information about branch lanes and intersections), and traffic information (e.g., information about the traffic state of the movement path to the potential MRC area (PMZ)) (e.g., information about severe traffic congestion, etc.), select a target MRC area (TMZ) in the potential MRC area (PMZ).

[0158] As an example, when calculating multiple potential MRC areas (PMZ), the processor 130 may use a pre-stored algorithm to determine the optimized potential MRC area (PMZ) among the multiple potential MRC areas (PMZ) as the target MRC area (TMZ) based on at least one of information about the size of the potential MRC area (PMZ), information about the distance to the potential MRC area (PMZ), information about the movement path to the potential MRC area (PMZ), information about the state of the vehicle 100, information about the number of lane changes, surrounding environment information, and traffic information. The above determination of the target MRC area (TMZ) in the potential MRC area (PMZ) is an example, and the determination is not limited thereto.

[0159] In operation S1140, when determining the target MRC zone (TMZ), the processor 130 may control the vehicle 100 to move into the target MRC zone (TMZ). When determining the target MRC zone (TMZ), the processor 130 may perform controls such as acceleration, deceleration, and lane change so that the vehicle 100 moves into the determined target MRC zone (TMZ).

[0160] Figure 14 is a flowchart showing operations for searching for potential MRC zones and determining a target MRC zone during shoulder parking during autonomous driving according to various embodiments of the present disclosure. Figure 13 The vehicle described in Figure 1 may be the vehicle 100 of

[0161] Refer to Figure 14 , as an example, when no potential MRC zone (PMZ) is searched within the ROI, the processor 130 may change the MRM type. The processor 130 may determine whether in-lane parking of type 2 is feasible and may perform in-lane parking when in-lane parking is feasible. When in-lane parking is not feasible, the processor 130 may determine whether straight parking of type 1 is feasible, and when straight parking is feasible, the processor 130 may perform straight parking. When straight parking is not feasible, the processor 130 activates the fault mitigation system (FMS).

[0162] Figure 14 is a flowchart showing operations for searching for potential MRC zones and determining a target MRC zone during shoulder parking during autonomous driving according to various embodiments of the present disclosure.

[0163] Refer to Figure 14 , when no target MRC zone (TMZ) is determined in the potential MRC zone (PMZ), the processor 130 may re-search the potential MRC zone (PMZ).

[0164] As an example, the processor 130 may end the re-search of the potential MRC zone (PMZ) when a predetermined condition for preventing re-search of the potential MRC zone (PMZ) from repeating (e.g., exceeding a predetermined allowable number of re-searches, exceeding a predetermined allowable re-search time, approaching the boundary of the ROI or deviating from the boundary of the ROI) is met.

[0165] In operation S1151, the processor 130 may determine whether shoulder parking for MRM (minimum risk maneuver) (i.e., MRM-SS) can be continuously performed.

[0166] As an example, when parking or moving of the vehicle 100 is not feasible, the processor 130 may determine that the MRM-SS cannot be continuously executed. As an example, when it is determined during the execution of the MRM-SS that the target MRC area (TMZ) is an incorrect location for stopping the vehicle 100 (e.g., a situation where another vehicle is parked at this location), the processor 130 may determine that vehicle parking is not feasible. As an example, the processor 130 may determine that moving is not feasible when disorder (e.g., malfunction) is detected in the vehicle 100 while executing the MRM-SS.

[0167] When the vehicle 100 cannot park in or move to the target MRC area (TMZ), the processor 130 may re-search for a potential MRC area (PMZ) or change the MRM type.

[0168] As an example, when the vehicle 100 cannot park in the target MRC area (TMZ) when moving into the target MRC area (TMZ) (e.g., another vehicle is parked in the target MRC area (TMZ)), the processor 130 may re-search for a potential MRC area (PMZ). When a new target MRC area (TMZ) is determined, the processor 130 may control the vehicle 100 to move into the new target MRC area (TMZ).

[0169] As an example, when the vehicle 100 cannot move when moving into the target MRC area (TMZ) (e.g., disorder occurs in the vehicle), the processor 130 may change the MRM type.

[0170] As an example, when the MRM-SS can be continued (parking or moving of the vehicle 100 into the target MRC area (TMZ) is feasible), the vehicle 100 may determine that the minimum risk condition (MRC) is satisfied.

[0171] In operation S1160, the processor 130 may determine whether type 2 in-lane parking is feasible based on the vehicle state information. The processor 130 may determine whether type 2 in-lane parking is feasible based on the functions that operate normally and / or the functions that do not operate normally among the functions required for autonomous driving.

[0172] In operation S1161, when in-lane parking is feasible, the processor 130 may perform in-lane parking.

[0173] In operation S1162, the processor 130 may determine whether in-lane parking can be continued. When parking or moving of the vehicle 100 is not feasible, the processor 130 may determine that in-lane parking cannot be continued.

[0174] As an example, when it is determined that in-lane parking cannot be continued, the processor 130 may change the MRM type.

[0175] As an example, when in-lane parking can be continued (when the vehicle's stopping or moving is feasible), the processor 130 may determine that the MRC is satisfied.

[0176] In operation S1163, when in-lane parking is not feasible, the processor 130 may determine whether type 1 straight-line parking is feasible based on the functions that operate normally and / or the functions that do not operate normally among the functions required for autonomous driving.

[0177] When straight-line parking is feasible, the processor 130 may perform straight-line parking.

[0178] As an example, the processor 130 may activate a failure mitigation system when straight-line parking is not feasible.

[0179] In operation S1164, the processor 130 may determine whether straight-line parking can be continued. When the vehicle 100's stopping or moving is not feasible, the processor 130 may determine that straight-line parking cannot be continued.

[0180] As an example, when it is determined that straight-line parking cannot be continued, the processor 130 may activate a failure mitigation system.

[0181] As an example, when it is determined that straight-line parking can be continued (when stopping or moving is feasible), the processor 130 may determine that the MRC is satisfied.

[0182] As described above, according to various embodiments of the present disclosure, the road shoulder of a road can be quickly searched and an optimized position for shoulder parking can be determined.

[0183] As described above, according to various embodiments of the present disclosure, when the search and determination of the position for shoulder parking fail, the failure can be quickly responded to through a re-search process or an MRM type change process.

Claims

1. An autonomous vehicle, comprising: At least one sensor configured to detect the surrounding environment of the vehicle and generate surrounding environment information; A processor configured to monitor the state of the vehicle to generate vehicle state information, and during autonomous driving of the vehicle, determine whether a minimum risk maneuver is required based on at least one of the surrounding environment information or the vehicle state information; And A controller configured to control the operation of the vehicle according to the control of the processor, Wherein, the processor is configured to: When it is determined that the minimum risk maneuver is required and it is determined that the roadside parking type is the type of minimum risk maneuver, set a region of interest; Search for a potential minimum risk condition (MRC) region, the potential MRC region being a candidate MRC region for roadside parking within the region of interest; and In the potential MRC region, select a target MRC region where the vehicle will park.

2. The autonomous vehicle according to claim 1, Among them, The region of interest has a rectangular shape including an upper side, a lower side, a left side, and a right side, Wherein, the upper side has a length from a point at the front of the vehicle to a position spaced apart from the front of the vehicle by a predetermined first distance, and Wherein, the left side has a length from a point on the center line to the boundary of the roadside.

3. The autonomous vehicle according to claim 1, Among them, The processor searches for the potential MRC region based on at least one of pre-stored map information, the surrounding environment information, and traffic information.

4. The autonomous vehicle according to claim 1, Among them, The processor determines, within the region of interest, a region that the vehicle can stop at by decelerating and moving to a lane different from the current position of the vehicle, is not occupied by an obstacle, and is larger than a predetermined reference size, as the potential MRC region.

5. The autonomous vehicle according to claim 1, Among them, The processor searches for the potential MRC region for a predetermined time.

6. The autonomous vehicle according to claim 1, Among them, After the search for the potential MRC region is completed, the processor continues to search for the potential MRC region in the background.

7. The autonomous vehicle according to claim 1, Among them, The potential MRC region includes a plurality of candidate MRC regions, and Wherein, each of the candidate MRC regions has a different size.

8. The autonomous vehicle according to claim 7, Among them, A gap region is formed between the candidate MRC region near the obstacle among the plurality of candidate MRC regions and the obstacle.

9. The autonomous vehicle according to claim 1, Among them, Based on an obstacle located in the roadside of the road, the potential MRC region includes: a first candidate MRC region located in the forward direction of the obstacle as the direction of the vehicle approaching; and a second candidate MRC region, based on the obstacle, located in the backward direction of the obstacle as the direction away from the vehicle. Wherein, a first gap region is formed between the first candidate MRC region and the obstacle, and wherein, a second gap region is formed between the obstacle and the second candidate MRC region.

10. The autonomous vehicle according to claim 9, Among them, the second gap region is larger than the first gap region.

11. The autonomous vehicle according to claim 1, Among them, the processor selects the target MRC region from among the potential MRC regions based on at least one of information about the size of the potential MRC region, information about the distance to the potential MRC region, information about the movement path to the potential MRC region, information about the state of the vehicle, information about the number of lane changes, the surrounding environment information, and traffic information.

12. The autonomous vehicle according to claim 1, Among them, when selecting the target MRC region, the processor controls the vehicle to move into the target MRC region.

13. A method for operating an autonomous vehicle, the method comprising: when, during autonomous driving of the vehicle, it is determined based on at least one of surrounding environment information or vehicle state information that a minimum risk maneuver is required and a road shoulder parking type is determined as the type of the minimum risk maneuver, setting an area of interest; searching for potential minimum risk condition MRC regions, where the potential MRC regions are candidate MRC regions for road shoulder parking within the area of interest; and selecting, from among the potential MRC regions, the target MRC region where the vehicle will park.

14. The method for operating an autonomous vehicle according to claim 13, Among them, searching for the potential MRC regions includes searching for the potential MRC regions based on at least one of pre-stored map information, the surrounding environment information, and traffic information.

15. The method for operating an autonomous vehicle according to claim 13, Among them, searching for the potential MRC regions includes: within the area of interest, determining an area that the vehicle can stop at by decelerating and moving from the current position of the vehicle to a different lane, is not occupied by an obstacle, and is larger than a predetermined reference size, as the potential MRC region.

16. The method for operating an autonomous vehicle according to claim 13, Among them, searching for the potential MRC regions includes searching for the potential MRC regions for a predetermined time.

17. The method for operating an autonomous vehicle according to claim 13, Among them, searching for the potential MRC regions includes: after completing the search for the potential MRC regions, continuing to search for the potential MRC regions in the background.

18. The method for operating an autonomous vehicle according to claim 13, Among them, Selecting the target MRC region includes: selecting the target MRC region from the potential MRC regions based on at least one of information about the size of the potential MRC region, information about the distance to the potential MRC region, information about the movement path to the potential MRC region, information about the vehicle state, information about the number of lane changes, the surrounding environment information, and traffic information.

19. The method for operating an autonomous vehicle according to claim 13, further comprising: When the target MRC region is selected, controlling the vehicle to move into the target MRC region.

Images