Vehicle executing minimum risk strategy and method of operating same

Abstract

The invention relates to a vehicle executing a minimum risk strategy and a method of operating the vehicle. A vehicle is disclosed that includes at least one sensor, a controller configured to control operation of the vehicle, and a processor electrically connected to the at least one sensor and the controller, where the processor is configured to: monitor vehicle status information and / or ambient information; determining whether automatic driving can be performed based on the vehicle state information and / or the surrounding environment information; executing a minimum risk strategy (MRM) when automatic driving cannot be performed; when a dangerous behavior is predicted even if MRM is performed, MRM is performed by additionally utilizing asymmetric braking that applies a braking force deviation between a left wheel and a right wheel of a vehicle.

Description

Technical Field

[0001] The present invention relates to a vehicle that implements a minimal risk strategy and a method of operating the vehicle. Background Art

[0002] Recently, advanced driver assistance systems (ADAS) have been under development to assist drivers in driving. ADAS has multiple sub-technical categories and provides convenience to drivers. Such ADAS is also referred to as automated driving or an automated driving system (ADS).

[0003] Meanwhile, when a vehicle performs automated driving, the automated driving system may malfunction. If an appropriate response is not made to such a malfunction of the automated driving system, the vehicle may be in a dangerous state. Summary of the Invention

[0004] Therefore, an object of an embodiment of the present invention is to solve the above disadvantages of the prior art. Embodiments of the present invention can provide a vehicle that implements a minimal risk maneuver (MRM) and a method of operating the vehicle to eliminate (or reduce) risks detected when normal automated driving cannot be performed during automated driving.

[0005] When a situation in which normal automated driving cannot be performed is detected during automated driving of a vehicle, an embodiment of the present invention can provide a control method configured to predict a dangerous behavior and perform braking when attempting to stop in a straight line by a minimal risk operation, and to mitigate or eliminate the dangerous behavior by performing asymmetric braking that applies a braking force deviation between the left and right wheels of the vehicle when the dangerous behavior occurs.

[0006] Aspects according to the present invention are not limited to the above aspects, and other aspects and advantages not mentioned above can be clearly understood from the following description and can be more clearly understood through the embodiments described herein.

[0007] According to an embodiment, a vehicle may include: at least one sensor; a controller configured to control the operation of the vehicle; and a processor electrically connected to the at least one sensor and the controller. The processor may be configured to: monitor vehicle state information and / or surrounding environment information; determine whether automated driving can be performed based on the vehicle state information and / or the surrounding environment information; when automated driving cannot be performed, perform a minimal risk maneuver (MRM); when a dangerous behavior is predicted even after performing the MRM, perform the MRM by additionally using asymmetric braking that applies a braking force deviation between the left and right wheels of the vehicle.

[0008] The processor may be further configured to: predict a situation where straight-line parking is selected as the type of MRM and the vehicle is traveling on a curved road as a dangerous behavior.

[0009] The processor may be configured to: predict a situation where straight-line parking is selected as the type of MRM due to partial or full non-execution of lateral control and the vehicle is traveling on a curved road as a dangerous behavior.

[0010] The processor may be configured to: predict a situation where lane departure, guardrail collision, or collision with another vehicle is predicted as a dangerous behavior.

[0011] The processor may be further configured to: when straight-line parking is selected as the type of MRM, set the MRM allowable space by using the longitudinal allowable distance and the lateral allowable distance; predict the stoppable distance, the collision risk distance, and the lane departure distance, and determine the shortest distance among the predicted stoppable distance, the predicted collision risk distance, and the predicted lane departure distance as the longitudinal allowable distance; determine that the lateral allowable distance is the smaller value between the value calculated by (adjacent lane width (W) - width of another vehicle traveling in the adjacent lane (Wv) - 0.75 m) and the maximum intrusion allowable range.

[0012] The processor may be configured to: predict a situation where the longitudinal allowable distance is shorter than the stoppable distance as a dangerous risk behavior.

[0013] When a dangerous behavior is predicted, the processor may be further configured to: predict at least one driving trajectory based on at least one braking force deviation; when the predicted driving trajectory based on the braking force deviation shows that the vehicle stops within the MRM allowable space, determine the braking force deviation to be applied to the vehicle from the at least one braking force deviation.

[0014] The processor may be further configured to: update the MRM allowable space when the vehicle turns due to asymmetric braking; re-determine the braking force deviation based on the updated MRM allowable space.

[0015] The processor may be further configured to: determine the braking force deviation to be applied to the vehicle by using artificial intelligence of reinforcement learning based on the driving trajectory information caused by the braking force deviation obtained through simulation or actual testing.

[0016] The processor may be further configured to: provide information about the vehicle speed, the preset MRM allowable space, and information about the curvature of the current driving lane as inputs for the artificial intelligence to make a determination.

[0017] According to another embodiment of the present invention, a method of operating a vehicle may include: monitoring vehicle state information and / or surrounding environment information; determining whether autonomous driving can be performed based on the vehicle state information and / or the surrounding environment information; when autonomous driving cannot be performed, executing MRM; predicting dangerous behaviors.

[0018] Executing MRM may include: when a dangerous behavior is predicted, executing MRM by additionally using asymmetric braking that applies a braking force deviation between the left and right wheels of the vehicle.

[0019] Predicting dangerous behaviors may include: predicting a situation where straight-line parking is selected as the type of MRM and the vehicle is traveling on a curved road as a dangerous behavior.

[0020] Predicting dangerous behaviors may include: predicting a situation where straight-line parking is selected as the type of MRM due to non-execution of lateral control partially or entirely and the vehicle is traveling on a curved road as a dangerous behavior.

[0021] Predicting dangerous behaviors may include: predicting a situation where lane departure, guardrail collision, or collision with another vehicle is predicted as a dangerous behavior.

[0022] Determining the type of MRM may select straight-line parking, and executing MRM may further include setting an MRM allowable space. Setting the MRM allowable space may include: predicting a stoppable distance, a collision risk distance, and a lane departure distance; determining the shortest distance among the predicted stoppable distance, the predicted collision risk distance, and the predicted lane departure distance as the longitudinal allowable distance; determining the smaller value between the value calculated by (adjacent lane width (W) - width of another vehicle traveling in the adjacent lane (Wv) - 0.75 m) and the maximum intrusion allowable range as the lateral allowable distance; setting the MRM allowable space based on the longitudinal allowable distance and the lateral allowable distance.

[0023] Predicting dangerous behaviors may further include: predicting a situation where the longitudinal allowable distance is shorter than the stoppable distance as a dangerous behavior.

[0024] When a dangerous behavior is predicted, executing MRM by additionally using asymmetric braking may further include: predicting at least one driving trajectory based on at least one braking force deviation; when the predicted driving trajectory based on the braking force deviation shows that the vehicle stops within the MRM allowable space, determining the braking force deviation to be applied to the vehicle from among the at least one braking force deviation.

[0025] When a dangerous behavior is predicted, executing MRM by additionally using asymmetric braking may further include: when the vehicle turns due to asymmetric braking, updating the MRM allowable space; re-determining the braking force deviation based on the updated MRM allowable space.

[0026] Determining the braking force deviation to be applied to a vehicle may include: determining the braking force deviation to be applied to the vehicle by using artificial intelligence of reinforcement learning based on the driving trajectory information caused by the braking force deviation obtained through simulation or actual testing.

[0027] Determining the braking force deviation to be applied to the vehicle may further include: providing information about the vehicle speed, a preset MRM tolerance space, and information about the curvature of the current driving lane as inputs for the artificial intelligence to make the determination.

[0028] According to an embodiment of the present invention, when a situation where normal autonomous driving cannot be performed is detected during autonomous driving, a minimum risk strategy operation may be executed, and the vehicle risk may be minimized, thereby improving safety. In particular, when performing the minimum risk strategy operation, if a dangerous behavior is predicted, additional asymmetric braking may be utilized to mitigate or eliminate the dangerous behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a block diagram of a vehicle according to an embodiment of the present invention;

[0030] Figure 2 is a schematic diagram showing the types of minimum risk operations according to an embodiment of the present invention;

[0031] Figure 3 is a flowchart showing the operation of a vehicle according to an embodiment of the present invention;

[0032] Figure 4 is a flowchart showing the vehicle stopping based on the minimum risk strategy;

[0033] Figure 5a 、 Figure 5b 、 Figure 6a and Figure 6b show an example of predicting the minimum risk strategy tolerance space;

[0034] Figure 7a 、 Figure 7b and Figure 7c show an example of eliminating a dangerous behavior situation based on asymmetric braking and parking within the tolerance space;

[0035] Figure 8a 、 Figure 8b and Figure 8c show an example of changing the tolerance space and eliminating a dangerous behavior based on asymmetric braking; and

[0036] Figure 9 shows an example of an artificial intelligence based on reinforcement learning for determining the braking force deviation of a vehicle. DETAILED DESCRIPTION

[0037] A detailed description will now be given with reference to the accompanying drawings according to the exemplary embodiments disclosed herein.

[0038] For a brief description with reference to the accompanying drawings, the same or equivalent components may be provided with the same reference numerals, and their descriptions will not be repeated.

[0039] Before describing the present invention in detail, the terms used therein may be defined as follows.

[0040] A vehicle is a vehicle equipped with an autonomous driving system (i.e., ADS) to enable autonomous driving. For example, the vehicle can perform at least one of steering, accelerating, decelerating, lane changing, and parking through the ADS without driver control. For example, the ADS may include at least one of the following: Pedestrian Detection and Collision Mitigation System (PDCMS), Lane Change Decision Aid System (LCDAS), Lane Departure Warning System (LDWS), Adaptive Cruise Control (ACC), Lane Keeping Assistance System (LKAS), Road Boundary Departure Prevention System (RBDPS), Curve Speed Warning System (CSWS), Forward Vehicle Collision Warning System (FVCWS), and Low Speed Following (LSF).

[0041] A driver refers to a person who uses a vehicle equipped with an ADS.

[0042] Vehicle control right is the authority to control at least one component of the 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, accelerating, decelerating (or braking), lane changing, lane detection, lateral control, obstacle recognition and distance detection, power system control, safety area detection, engine on / off, power on / off, and vehicle locking / unlocking functions. The above functions are examples to help understand the embodiments, but the embodiments of the present invention are not limited thereto.

[0043] The shoulder represents the space between the outermost road boundary (or the outermost lane boundary) in the direction of vehicle travel and the road edge (e.g., curb and guardrail). That is, the shoulder is a part of the road provided at the road edge, which can represent the space where emergency vehicles can park, emergency vehicles can bypass traffic jams, or vehicles can leave traffic jams.

[0044] Figure 1 is a block diagram of a vehicle according to an embodiment of the present invention. Figure 2 is a schematic diagram showing the minimum risk operation type according to an embodiment of the present invention.

[0045] Figure 1 The vehicle configuration shown is an embodiment, and each component can be composed of a chip, a component, an electronic circuit, or a combination of chips, components, and / or electronic circuits. According to one embodiment, Figure 1 Some of the components shown can be separate multiple components, configured as different chips, different components, or different electronic circuits, some of which can be combined into one chip, one component, or one electronic circuit. According to one embodiment, some Figure 1 of the components shown in can be omitted, or other components not shown in the figure can be added. Figure 1 At least some of the components shown can be described with reference to the following drawings.

[0046] Referring to Figure 1 , vehicle 100 can include a sensor unit 110, a controller 120, a processor 130, a display 140, and a communication device 150.

[0047] According to an embodiment, the sensor unit 110 can be configured to detect the surrounding environment of the vehicle 100 by using at least one sensor and generate data related to the surrounding environment based on the detection result.

[0048] For example, the sensor unit 110 can obtain information about objects near the vehicle (e.g., another vehicle, a person, an object, a curb, a guardrail, a lane, and an obstacle) based on the sensing data obtained by at least one sensor. The information about the objects near the vehicle can include at least one of the position, size, and shape of the object, the distance to the object, and the relative speed of the object.

[0049] As another example, the sensor unit 110 may measure the position of the vehicle 100 by using at least one sensor. For example, the sensor unit 110 may include at least one of 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. These sensors are examples to help understand the embodiments, and the sensors of the present invention are not limited thereto.

[0050] According to one embodiment, the camera may be configured to capture an image of the surrounding environment of the vehicle and generate image data including an object located in front of, behind, and / or to the side of the vehicle 100. According to one embodiment, the radar may generate information about an object located in front of, behind, and / or to the side of the vehicle 100 by using electromagnetic waves (or radio waves). According to one embodiment, the ultrasonic sensor may generate information about an object located in front of, behind, and / or to the side of the vehicle 100 by using ultrasonic waves. According to one embodiment, the infrared sensor may generate information about an object located in front of, behind, and / or to the side of the vehicle 100 by using infrared rays.

[0051] According to one embodiment, the position measurement sensor may be configured to 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 position data of the vehicle based on a signal generated by at least one of the GPS sensor, the DGPS sensor, and the GNSS sensor.

[0052] According to an embodiment, the controller 120 may be configured to control the operation of at least one component and / or at least one function of the vehicle 100 based on the control of the processor 130. For example, 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, power system control, safety zone detection, engine on / off, power on / off, and vehicle lock / unlock functions.

[0053] According to one embodiment, for the autonomous driving and / or the minimum risk strategy (MRM) of the vehicle 100, the controller 120 may be configured to control the operation of at least one component and / or at least one function of the vehicle 100 based on the control of the processor 130. For example, for the operation of the minimum risk strategy, the controller 120 may be configured to control the operation of at least one of the functions of steering, acceleration, deceleration (or braking), lane change, lane detection, lateral control, obstacle recognition and distance detection, powertrain control, and safety zone detection.

[0054] According to an embodiment, the processor 130 may be configured to control the overall operation of the vehicle 100. According to one embodiment, the processor 130 may include an electronic control unit (ECU) configured to control the components of the vehicle 100 as a whole. For example, the processor 130 may include a central processing unit (CPU) or a micro processing unit (MCU) capable of performing computational processing.

[0055] When a specified event occurs, the processor 130 may control the components within the vehicle 100 to activate the ADS so that the vehicle can perform autonomous driving. The specified event may occur when a driver request for autonomous driving control of the vehicle is recognized, the driver delegates the vehicle control right, or a condition predefined by the driver and / or the vehicle designer is satisfied.

[0056] The processor 130 may determine whether normal autonomous driving can be performed based on at least one of the vehicle state information and the surrounding environment information during autonomous driving. According to one embodiment, the processor 130 may monitor the mechanical and / or electrical states of the vehicle internal components (such as sensors, actuators, etc.) from the time when the ADS is activated, and obtain vehicle state information indicating whether mechanical defects and / or electrical defects of the vehicle internal components occur. The vehicle state information may include information about the mechanical state and / or electrical state of the vehicle internal components. For example, the vehicle state information may include information indicating whether the functions required for autonomous driving can operate normally based on the mechanical state and / or electrical state of the vehicle internal components. According to one embodiment, the processor 130 may obtain the surrounding environment information of the vehicle through the sensor unit 110 from the time when the ADS is activated.

[0057] The processor 130 may determine whether the functions required for autonomous driving can operate properly based on the vehicle state information. The functions required for autonomous driving may include 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 function, and a distance detection function. When at least one of the functions required for autonomous driving cannot be executed, the processor 130 may determine that normal autonomous driving cannot be performed.

[0058] The processor 130 may determine whether the vehicle state is suitable for normal 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 information about battery overheating and tire pressure) is suitable for normal driving conditions. When the vehicle state is not suitable for normal driving conditions, the processor 130 may determine that normal autonomous driving cannot be performed.

[0059] The processor 130 may determine whether the environment around the vehicle is suitable for the Operation Design Domain (ODD) of autonomous driving based on at least one of the surrounding environment information. When the surrounding environment information of the vehicle is not suitable for the ODD, the processor 130 may determine that normal autonomous driving cannot be performed.

[0060] When normal autonomous driving cannot be performed, the processor may determine that this is a situation that requires the execution of the Minimum Risk Maneuver (MRM) to minimize the accident risk. When it is a situation that requires the execution of the MRM, the processor 130 may determine the type of the MRM.

[0061] Even when the ADS system is working properly but an abnormal signal of the driver is detected or an emergency stop is required, the processor 130 may also determine that it is a situation that requires the execution of the MRM. The abnormal signal of the driver may include an abnormality in the driver's vital signs or a lack of response to the handover request. The emergency stop may be requested by the driver or a third party (such as the police).

[0062] Referring to Figure 2 showing the types of the Minimum Risk Maneuver (MRM) according to the embodiments, the types of the MRM may include in-lane stop 201, shoulder stop 203, and straight-line stop 205.

[0063] The in-lane stop 201 may include an in-lane parking type (type 1) configured to park within the boundary of the lane in which the vehicle is traveling; and a parking type with a unilateral lateral deviation (type 2) configured to park the vehicle when a part of the vehicle is allowed to deviate from the lane line in one direction.

[0064] For example, the type of in-lane parking 201 can represent the type in which the vehicle stops within the boundary of the lane in which the vehicle is traveling or partially outside the boundary of the lane based on lateral control and / or deceleration.

[0065] The types of shoulder parking 203 can include an in-shoulder parking type (type 3), a shoulder parking type with a lateral deviation (type 4), a combined parking type with bilateral lateral deviations (type 5), and a combined parking type with a longitudinal margin (type 6).

[0066] For example, the shoulder parking 203 can represent the following type, in which the vehicle moves such that part or all of the vehicle leaves the road boundary (or the boundary of the outermost lane) and is located on the shoulder through longitudinal acceleration, longitudinal deceleration, and / or lateral control and then stops.

[0067] The types of straight-line parking 205 can include a longitudinal parking type (type 7), in which the vehicle travels in a straight line rather than along the lane and then stops.

[0068] The straight-line parking 205 is a type in which the vehicle stops by only using the longitudinal deceleration function without using lateral control. For example, in a case where the lane detection function cannot be performed or lateral control cannot be performed due to a defect in the actuator for lateral control, the straight-line parking can be performed.

[0069] The processor 130 can determine the type of the minimum risk strategy as one of the in-lane parking 201, the shoulder parking 203, and the straight-line parking 205 based on at least one of the vehicle state information, the external environment information, and the road traffic regulations.

[0070] According to one embodiment, regardless of the vehicle state information, the external environment, and the road traffic regulations, the processor 130 can determine the type of the minimum risk strategy based on a preset basic type pre-specified by the designer. The pre-specified basic type can be the shoulder parking 203. This is because the road traffic regulations allow (or recommend) parking on the shoulder, and the shoulder is a relatively safe area in case of an emergency due to vehicle failure or other reasons. Therefore, in the embodiment of the present invention, the basic type can be pre-specified as the shoulder parking 203, so as to minimize the possibility of a secondary collision, minimize the impact of the minimum risk strategy of the vehicle on the traffic flow, and enable the driver or passengers to evacuate from the road.

[0071] According to one embodiment, when the ADS system is operating normally but the vehicle state is not suitable for normal driving conditions, the processor 130 can determine the type of the minimum risk strategy as the shoulder parking 203. For example, when the ADS is operating normally but battery overheating or a tire blowout is detected, the processor 130 can determine the type of the minimum risk strategy as the shoulder parking 203.

[0072] According to one embodiment, even when an abnormal signal of the driver is detected or an emergency stop is required, the processor 130 may determine the type of the minimum risk strategy as the roadside stop 203.

[0073] According to one embodiment, once the type of the minimum risk strategy is determined as the roadside stop 203, the processor 130 may control the position of the vehicle through lateral and / or longitudinal control to move the vehicle toward the roadside, such that at least a part of the vehicle may be located on the roadside.

[0074] According to one embodiment, if there is no roadside area within a specified threshold range where the vehicle can stop based on the current position of the vehicle 100, the processor 130 may change the type of the minimum risk strategy. For example, the processor 130 may change the type of the minimum risk strategy to the in-lane stop 201 and then perform the in-lane stop.

[0075] According to one embodiment, the processor 130 may determine whether the roadside stop is completed within a specified time, and determine whether to change the type of the minimum risk strategy based on the determination result. Unless the roadside stop is completed within the specified time, the processor 130 may change the type of the minimum risk strategy from the roadside stop to the in-lane stop 201. For example, when the roadside stop is not completed within the specified time, the processor 130 may change the type of the minimum risk strategy from the roadside stop 203 to the in-lane stop 201, and then control the vehicle to stop in the lane. When the in-lane stop is not completed within the specified time, the processor 130 may change the type of the minimum risk strategy from the in-lane stop to the straight-line stop 205 and then control the execution of the straight-line stop.

[0076] According to an embodiment, the processor 130 may perform an operation for stopping the vehicle based on the determined type of the minimum risk strategy, and determine whether a minimal risk condition (MRC) is satisfied. The MCR may indicate that the speed of the vehicle is 0 (zero). For example, when performing at least one operation based on the determined type of the minimum risk strategy, the processor 130 may determine whether the vehicle is in a stationary state with a speed of 0. When the vehicle 100 is in a stationary state with a speed of 0, the processor 130 may determine that the MRC is satisfied.

[0077] According to an embodiment, when the MCR is satisfied, the processor 130 may terminate the minimum risk strategy and convert the ADS to a standby mode or a shutdown state. According to one embodiment, the processor 130 may switch the ADS to a standby mode or a shutdown state, and then control the vehicle control right to transfer it to the driver (or user).

[0078] According to an embodiment, the display 140 may be configured to visually display information related to the vehicle. For example, the display 140 may provide diverse information related to the state of the vehicle 100 to the driver of the vehicle 100 based on the control of the processor 130. The diverse information related to the vehicle state may include at least one of information indicating whether various components of the vehicle and / or at least one function of the vehicle are operating normally and information indicating the driving state of the vehicle. For example, the driving state of the vehicle may include at least one of the state of an autonomous driving vehicle, the state in which the vehicle executes a minimal risk strategy, the state of completing the minimal risk strategy, and the state of terminating autonomous driving.

[0079] According to an embodiment, the communication device 150 may communicate with an external device of the vehicle 100. According to an embodiment, the communication device 150 may be configured to receive data from the outside of the vehicle 100 or transmit data to the outside of the vehicle 100 based on the control of the processor 130. For example, the communication device 150 may perform communication using a wireless communication protocol or a wired communication protocol.

[0080] As Figure 1 shown, the controller 120 and the processor 130 are separate components, but according to various embodiments, the controller 120 and the processor 130 may be integrally formed as one component.

[0081] Figure 3 is a flowchart showing the operation of a vehicle according to an embodiment of the present invention.

[0082] Refer to Figure 3 , in step S310, the vehicle 100 may enable the ADS to operate normally.

[0083] According to one embodiment, the vehicle 100 may monitor the vehicle state and the surrounding environment when performing autonomous driving based on the normal operation of the ADS. The vehicle 100 may monitor the vehicle and the surrounding environment and detect whether MRM is required based on the obtained information. When MRM is required, an A1 event may be generated.

[0084] According to one embodiment, the vehicle 100 may detect whether driver (or user) intervention is required when performing the ADS based on the normal operation of the ADS. If driver intervention is required, the vehicle 100 may execute an intervention request (RequestTo Intervene, RTI) or issue a warning through the ADS. The RTI or warning may be an A2 event. When an A1 event occurs in the state where the ADS is operating normally, the vehicle 100 may execute step S320.

[0085] When an A2 event occurs while the ADS is operating normally, vehicle 100 may perform RTI at step S350 and determine whether RTI is detected within a specified time. When RTI is not detected within the specified time, vehicle 100 may determine that a B1 event has occurred. When the B1 event has occurred, vehicle 100 may perform step S320. When RTI is detected within the specified time, vehicle 100 may determine whether a B2 event has occurred. When the B2 event has occurred, vehicle 100 may perform step S340.

[0086] At step S320, vehicle 100 may perform MRM. According to one embodiment, vehicle 100 may determine the type of MRM based on at least one of vehicle state information, surrounding environment information, and road traffic regulations. According to one embodiment, regardless of vehicle state information, external environment information, and road traffic regulations, vehicle 100 may determine the type of MRM as a basic type pre-specified by the designer. The type of MRM may be one of in-lane parking 201, shoulder parking 203, and straight parking 205 as Figure 2 shown.

[0087] Vehicle 100 may control at least one component provided therein to park based on the determined type of MRM. According to one embodiment, vehicle 100 may send information indicating that the vehicle is performing MRM to other vehicles.

[0088] At step S320, vehicle 100 may determine whether the minimal risk condition (MRC) is satisfied by performing MRM so as to reduce the speed of the vehicle to 0 (zero). When the MRC is satisfied, vehicle 100 may determine that a C1 event has occurred and perform step S330. Vehicle 100 may determine whether driver intervention is detected during MRM. When driver intervention is detected, vehicle 100 may determine that a C2 event has occurred and perform step S340.

[0089] Vehicle 100 may maintain the state where the MRC is satisfied. The state where the MRC is satisfied may represent a state where the vehicle is stationary. For example, vehicle 100 may maintain the state where the vehicle is stationary. For example, vehicle 100 may perform a control operation configured to maintain the stationary state of the vehicle regardless of the road surface inclination at the stop position. When maintaining the state where the MRC is satisfied, vehicle 100 may determine whether a D1 event occurs. The D1 event may include at least one of the driver turning off the ADS and transferring vehicle control to the driver. When the D1 event has occurred, vehicle 100 may perform step S340.

[0090] In step S340, vehicle 100 may switch the ADS to the standby mode or the off state. Vehicle 100 does not execute the ADS during the standby mode or the off state.

[0091] The above steps S310, S320, S330, and S350 are the states where the ADS is activated, and step S340 is the state where the ADS is not activated.

[0092] Hereinafter, the operation of vehicle 100 to perform MRM in S320 will be described in detail. In particular, the operation of performing MRM when the type of MRM is a straight-line parking will be described in detail.

[0093] Figure 4 is a flowchart showing the vehicle stopping based on the minimum risk strategy.

[0094] Figure 4 The operation of can be Figure 3 The specific operation of S320 shown. In particular, this can be the operation of MRM when the type of MRM is determined to be a straight-line parking. Figure 4 Each step shown in the embodiment of can be executed sequentially, but is not limited to being executed sequentially. For example, the order of the steps can be changed, and at least two steps can be executed in parallel. In addition, Figure 4 The operation of can be executed by the processor 130 and / or the controller 120 provided in vehicle 100, or they can be implemented by instructions executable by the processor 130 and / or the controller 120.

[0095] Figure 4 The flowchart of shows that vehicle stopping is performed based on MRM, but when the vehicle is stationary according to the driver's request, even if this situation does not require MRM, it can be used equally and / or similarly Figure 4 The flowchart of.

[0096] Refer to Figure 4 In step S401, vehicle 100 may select the type of MRM.

[0097] According to one embodiment, based on the vehicle state information and / or the surrounding environment information, vehicle 100 may perform MRM. When it is determined that the ADS is working properly but the vehicle state is not suitable for normal driving conditions, vehicle 100 may perform MRM. In addition, even when an abnormal signal of the driver or an emergency stop requested by the driver is detected, vehicle 100 may perform MRM.

[0098] According to various embodiments of the present invention, the vehicle 100 may select straight parking as the type of MRM based on vehicle state information and / or surrounding environment information to perform MRM. According to one embodiment, when lateral control is not possible or a lane or a shoulder is detected when braking control is possible, the vehicle 100 may select straight parking as the type of MRM.

[0099] In step S403, the vehicle 100 may predict an MRM allowable space. The MRM allowable space may represent a space where the vehicle 100 may be stationary through the operation of the MRM.

[0100] Figure 5a , Figure 5b , Figure 6a and Figure 6b An example of predicting the allowed space of minimum risk strategies is shown.

[0101] refer to Figure 5a and Figure 5b , the vehicle can predict the MRM allowable spaces 510 and 520 based on the comprehensive information about the guardrail and the position / speed information of the front vehicle and the surrounding vehicles. In addition, the vehicle 100 can improve the prediction performance of the MRM allowable space by additionally considering the positioning information and the map information, and can improve the prediction performance of the MRM space by predicting the driving trajectories of the vehicle and the front vehicle. According to one embodiment, the prediction of the MRM allowable space can be performed by utilizing artificial intelligence.

[0102] 6, the vehicle 100 can predict the longitudinal allowable distance (D long ) and lateral allowable distance (D lat ), thereby predicting the MRM tolerance space.

[0103] According to one embodiment, the vehicle 100 can predict the parking distance (D S ), collision risk distance (D c ) and lane departure distance (D d ), thereby predicting the longitudinal allowable distance (D long ).

[0104] According to one embodiment, the stoppable distance (D ) may be predicted based on the current vehicle speed (v), the deceleration of the MRM (α), and a gap distance constant (δ) set for safety. S ), as shown in the following mathematical equation 1.

[0105] [Mathematical equation 1]

[0106]

[0107] When straight parking is selected as the type of MRM and there is a guardrail on the left side of the driving lane, for a vehicle 610 in lane 1 such as Figure 6a , if the lane is curved, there is a point where the driving trajectory of the vehicle 610 intersects with the guardrail, so that a collision risk point 620 can be selected. In addition, the vehicle 610 can more accurately select the collision risk point 620 by additionally considering the width of the vehicle. The distance within the driving trajectory from the current position to the selected collision risk point 620 can be defined as the collision risk distance (D c ).

[0108] According to another embodiment, when straight parking is selected as the type of MRM, if the lane is curved, then Figure 6a , a vehicle 630 on lane 2 can select a lane departure point 640 by creating a point where the predicted driving trajectory intersects with the lane. In addition, the vehicle 630 can more accurately select the lane departure point 640 by additionally considering the width of the vehicle.

[0109] Therefore, the distance from the current position to the lane departure point 640 can be defined as the lane departure distance (D d ).

[0110] The vehicle 100 can define the longitudinal allowable distance (D long ) as the minimum value among the stoppable distance D s , the collision risk distance D c , and the lane departure distance D d , i.e., min(D s , D c and D d ).

[0111] Referring to Figure 6b , the lane width W and the maximum intrusion allowable range M t (e.g., 1 m) can be defined.

[0112] According to one embodiment, when parking based on MRM, the vehicle 650 may intrude into the adjacent lane. For example, even if straight parking is selected as the type of MRM, if the steering is towards the adjacent lane, it is possible to encroach on the adjacent lane because even if only one tire bursts and the vehicle moves forward, the vehicle will gradually move laterally. In addition, even if in-lane parking is selected as the type of MRM, the vehicle will park within the lane with the condition of allowing intrusion into the adjacent lane.

[0113] In this case, the adjacent lane intrusion allowable range M tIt is necessary to enable vehicles traveling in adjacent lanes to avoid the stopped vehicle without changing lanes. That is, the avoidance allowance space for vehicles in adjacent lanes within the lane width W of the adjacent lane must be greater than the vehicle width in the adjacent lane + 0.75 m. Therefore, the adjacent lane intrusion allowance range M t can be the smaller value between the value obtained by W - Wv - 0.75 m (where W is the width of the adjacent lane and Wv is the width of the vehicle traveling in the adjacent lane) and the maximum intrusion allowance range (e.g., 1 m).

[0114] The lateral allowance distance D lat can be determined based on the position of the vehicle 100 within the lane. When the center of the vehicle 100 is located at X lat meters from the left starting point of the driving lane, the left lateral allowance distance (D al,l ) of the vehicle 100 can be predicted as (X lat - Wv / 2 + M t ). Here, Wv represents the width of the vehicle, and a value can be set for each vehicle type (small, medium, and large), or a value can be set for all vehicles. M t is the adjacent lane intrusion allowance range obtained above.

[0115] The vehicle 100 can predict the MRM allowance space based on the longitudinal allowance distance (D long ), the left lateral allowance distance (D lat,l ), and the right lateral allowance distance (D lat,r ).

[0116] Referring again to Figure 4 , in step S405, the vehicle can be controlled to stop within the predicted allowance space.

[0117] According to one embodiment, when the longitudinal allowance distance (D long ) is determined to be the stoppable distance D s , the vehicle 100 can decelerate at a pre-specified deceleration. According to one embodiment, the specified deceleration can be set to the deceleration of the MRM (e.g., -4 m / s 2 ).

[0118] According to one embodiment, when the longitudinal allowance distance D long is determined to be the collision risk distance D c or the lane departure distance D d , in the case of decelerating at the specified deceleration, the vehicle may not be able to stop within the allowance space. Therefore, additional vehicle control operations are required to make the vehicle stop within the allowance space.

[0119] According to one embodiment, vehicle 100 can change the deceleration start point. For example, when the MRM is triggered, the taillights can be set to flash for 3 seconds and then deceleration starts. However, in order to stop within the allowable space, vehicle 100 can start decelerating while triggering the MRM. Additionally, once the MRM is triggered and the calculation of the allowable space is completed, deceleration can start.

[0120] According to another embodiment, vehicle 100 can adjust the deceleration to stop within the allowable space. For example, a specified deceleration can be set as the deceleration of the MRM (e.g., -4m / s 2 ), but the absolute value of the deceleration can be increased to reduce the braking distance by decelerating faster.

[0121] However, when considering that a sudden stop with an increased absolute value of deceleration increases the risk of collision with the vehicle behind, controlling the deceleration start point can have a higher priority than controlling the deceleration. Additionally, according to one embodiment, controlling the deceleration start point and controlling the deceleration / acceleration can also be combined and executed simultaneously.

[0122] According to another embodiment, if vehicle 100 does not stop within the allowable space based on the normal MRM, additional asymmetric braking can be utilized. For example, when a dangerous behavior is predicted during the MRM, vehicle 100 can utilize asymmetric braking to mitigate or eliminate the dangerous behavior situation. For example, the dangerous behavior situation can be a situation where vehicle 100 cannot stop within the allowable space based only on the current MRM. A dangerous behavior situation may occur when vehicle 100 performs a straight stop type of MRM while driving on a curved road, for example, a situation where the vehicle crosses lanes and collides with a vehicle in another lane or collides with a guardrail. According to another embodiment, when the lateral control of vehicle 100 is not fully or partially performed by the processor, if the lateral control of vehicle 100 is not partially or fully operated by the processor, a dangerous behavior situation may occur, that is, the vehicle crosses lanes and collides with a vehicle in another lane because it does not perform the lateral control as required by the processor.

[0123] In the above dangerous behavior situations, the vehicle can mitigate or eliminate the dangerous behavior situation by utilizing asymmetric braking.

[0124] Asymmetric braking control is a technique that generally generates a turning moment by applying a braking force difference between the left and right wheels, thereby turning the vehicle.

[0125] Figure 7a 、 Figure 7b and Figure 7c shows an example of eliminating the dangerous behavior situation based on asymmetric braking and stopping within the allowable space.

[0126] ReferenceFigure 7a , if asymmetric braking is not performed, there may be a dangerous behavior situation where the vehicle 100 travels in a straight line, exceeds the longitudinal allowable distance, exits the allowable space 710, and reaches the MRC state. That is, when another vehicle or a wall is within the longitudinal allowable distance D long , such a dangerous behavior situation may occur: that is, the vehicle 701 reaches the MRC state outside the allowable space 710 just by traveling in a straight line along the longitudinal allowable distance D long . That is, when another vehicle or a wall is within the longitudinal allowable distance D long , a dangerous behavior situation of collision may occur. As Figure 7b shown, when asymmetric braking is applied, the vehicle 703 turns laterally and moves laterally to reach the MRC state. Even in Figure 7b , the vehicle cannot reach the MRC state within the complete allowable space 710 and may reach the MRC state after exceeding the longitudinal allowable distance D long . However, dangerous behavior situations can be eliminated, such as weakening the collision intensity or changing the collision part. When asymmetric braking is applied with a greater force in Figure 7c than in Figure 7b , the vehicle 705 can travel further in the lateral direction to eliminate the dangerous behavior situation and reach the MRC state within the allowable space 710.

[0127] Although Figure 7a , Figure 7b and Figure 7c are not shown, if too much asymmetric braking is applied, the vehicle may dangerously turn in the turning direction or move beyond the lateral allowable distance (e.g., D lat,l or D lat,r ), thus exiting the allowable space 710.

[0128] Therefore, an appropriate amount of asymmetric braking needs to be performed so that the vehicle 100 can stop within the allowable space 710.

[0129] For this purpose, the vehicle 100 can predict the driving trajectory for each at least one braking force deviation. The vehicle 100 can determine the braking force deviation based on determining whether there is a driving trajectory in at least one predicted driving trajectory that enables the vehicle to reach the MRC state within the allowable space 710. According to one embodiment, the vehicle 100 can predict the driving trajectory by gradually increasing the braking force deviation or gradually decreasing the braking force deviation. Or, the vehicle 100 can arbitrarily select a plurality of braking force deviations and predict the driving trajectory for the selected plurality of braking force deviations to determine an appropriate braking force deviation. According to another embodiment, the vehicle 100 can select a plurality of braking force deviations based on past information.

[0130] Figure 8a, Figure 8b and Figure 8c shows an example of changing the allowable space based on asymmetric braking and eliminating dangerous behavior.

[0131] Refer to Figure 8a , in the case of a curved lane, when vehicle 800 performs a straight-line stop as the MRM type, as Figure 8a shown, since the first allowable space 810 for stopping the vehicle in the MRC state is short, it is possible to predict the dangerous behavior of colliding with the first collision risk point 811 of the guardrail, as Figure 8a shown. When vehicle 801 performs asymmetric braking in this case, vehicle 801 can turn, and then the collision risk point of vehicle 801 changes from the first collision risk point 811 to the second collision risk point 821, and the allowable space can change from the first allowable space 810 to the second allowable space 820. For this reason, the longitudinal allowable distance can be long enough to eliminate the dangerous behavior situation.

[0132] In addition, when vehicle 801 continuously maintains the current asymmetric braking or increases the braking force deviation between the two wheels, the longitudinal allowable distance can be the third collision risk distance based on the third collision risk point 831. Since the third collision risk distance is longer than the stoppable distance of vehicle 801, the dangerous behavior situation can be eliminated. Then, vehicle 801 can terminate the asymmetric braking and control the preset acceleration / deceleration of the MRM to reach the MRC state within the allowable space.

[0133] Vehicle 801 can predict the MRM allowable space in real time when performing asymmetric braking. If the MRM allowable space is determined by the stoppable distance based on the prediction result, the vehicle can determine that the dangerous behavior situation can be eliminated and does not perform additional asymmetric braking. Finally, the vehicle can be controlled to stop within the finally determined MRM allowable space.

[0134] According to another embodiment, vehicle 100 can determine the braking force deviation based on artificial intelligence of reinforcement learning.

[0135] Figure 9 shows an example of artificial intelligence of reinforcement learning for determining the braking force deviation of a vehicle.

[0136] Refer to Figure 9 , the artificial intelligence 910 of reinforcement learning can be composed of a convolutional layer 913 and a fully connected (FC) layer 915, but the embodiment is not limited thereto.

[0137] The artificial intelligence 910 for reinforcement learning may be an artificial intelligence that performs reinforcement learning by utilizing driving trajectory information based on a braking force deviation obtained through simulation or actual testing. According to another embodiment, the artificial intelligence 910 for reinforcement learning may be an artificial intelligence that performs reinforcement learning by utilizing driving trajectory information based on the initial speed and braking force of the vehicle. According to yet another embodiment, the artificial intelligence 910 for reinforcement learning may learn to obtain Figure 7c or Figure 8c the results shown.

[0138] As described above, the artificial intelligence 910 for reinforcement learning may output information on the braking force deviation that the vehicle should execute by inputting information on the vehicle speed, the determined allowable space, and the curvature of the current driving lane. Here, the input of the artificial intelligence 910 may at least include the vehicle speed, the determined allowable space, and the curvature information of the current driving lane, but the input is not limited thereto.

[0139] The vehicle may be controlled to apply a braking force deviation based on the braking force deviation information output from the artificial intelligence 910.

[0140] As described above, in various embodiments of the present invention, a vehicle control method is proposed to predict the allowable space when selecting straight parking as the MRM type, so that the vehicle stops within the allowable space. According to one embodiment, the vehicle may perform at least one of vehicle control operations such as an early deceleration start point, increasing the absolute value of the deceleration, or performing asymmetric braking, so that the vehicle stops within the allowable space.

[0141] When the vehicle turns due to asymmetric braking, the vehicle that performs asymmetric braking may update the allowable space in real time, and then the vehicle may finally stop within the allowable space.

[0142] Through this control, when the vehicle 100 needs MRM, the vehicle 100 may safely stop within the allowable space.

[0143] In one or more exemplary embodiments, the above functions may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media includes both communication media and computer storage media, which includes any medium that facilitates the transfer of a computer program from one place to another. The storage media may be any available media that can be accessed by a computer. By way of example and not limitation, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0144] When an embodiment is implemented as a program code or code segment, it should be recognized that the code segment may represent a procedure, function, subroutine, program, routine, subroutine, module, software package, class, or any combination of instructions, data structures, or program statements. By passing and / or receiving information, data, arguments, parameters, or memory contents, the code segment may be coupled to other code segments or hardware circuits. Information, arguments, parameters, data, etc. may be transferred, sent, or transmitted in any suitable manner, including memory sharing, message passing, token passing, network transmission, etc. Additionally, in some aspects, the steps and / or operations of a method or algorithm may exist as one or any combination of code and / or instructions, or a set of code and / or instructions, on a machine-readable medium and / or computer-readable medium, which may be incorporated into a computer program product.

[0145] In a software implementation, the techniques described herein may be implemented as modules (e.g., procedures, functions, etc.) that perform the functions described herein. The memory unit may be implemented inside or outside the processor, in which case the memory unit may be communicatively coupled to the processor in a variety of well-known ways.

[0146] In a hardware implementation, the processing unit may be implemented in one or more of the following: one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or combinations thereof.

[0147] The foregoing includes examples of one or more embodiments. Of course, it is not possible to describe every possible combination of components or methods for purposes of describing the above embodiments, but one of ordinary skill in the art will recognize that many additional combinations and permutations of the various embodiments are possible. Accordingly, the described embodiments are intended to embrace all such alternatives, modifications, and variations that fall within the true spirit and scope of the appended claims. Further, if the term "comprising" is used in a specific embodiment or claim, such term is intended to be inclusive in a manner similar to the term "consisting of" as interpreted when employed as a transitional word in a claim.

[0148] As used herein, the term "infer" or "inference result" generally refers to the process of making a determination or inference about the state of a system, environment, and / or user based on a set of observations captured via events and / or data. For example, an inference result can be used to identify a particular situation or action, or can generate, for example, a probability distribution over states. An inference result can be probabilistic, i.e., it can be based on an examination of data and events to compute a probability distribution over relevant states. Such an inference result enables us to infer new events or behaviors based on a set of observed events and / or stored event data, whether or not the events are closely related in time and whether or not the events and data are from one or more events and data sources.

[0149] In addition, as used in this application, the terms "component", "module", "system", etc. include, but are not limited to, computer-related entities such as hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable thread, a program, and / or a computer. For example, an application running on a computing device and the computing device can both be components. One or more components can exist in a process and / or an execution thread, and the components can be concentrated on one computer and / or distributed between two or more computers. In addition, these components can execute from various computer-readable media storing various data structures. The components can communicate through local and / or remote processes, such as by signals having one or more data packets (e.g., data from a local system, from other components in a distributed system, and / or from components interacting with other systems through a network such as the Internet).

Claims

1. A vehicle comprising: at least one sensor; a controller configured to control operation of the vehicle; as well as a processor electrically connected to the at least one sensor and the controller; Wherein, the processor is configured as follows: Monitoring vehicle status information and / or surrounding environment information; Determining whether autonomous driving is possible based on vehicle state information and / or surrounding environment information; When autonomous driving is not possible, implement the minimum risk strategy; When a dangerous behavior is predicted even though MRM is executed, a minimum risk strategy is executed by additionally utilizing asymmetric braking that applies a braking force bias between the left and right wheels of the vehicle.

2. The vehicle according to claim 1, wherein: The processor is further configured to: The situation where straight-line parking is selected as the type of the minimum risk strategy and the vehicle is traveling on a curved road is predicted as a dangerous behavior.

3. The vehicle according to claim 2, wherein: The processor is further configured to: A situation in which straight-line parking is selected as the type of minimum risk strategy because lateral control is not partially or completely performed and the vehicle is traveling on a curved road is predicted as a dangerous behavior.

4. The vehicle according to claim 2, wherein: The processor is further configured to: A lane departure, guardrail collision, or collision with another vehicle is predicted as a dangerous behavior.

5. The vehicle according to claim 1, wherein: The processor is further configured to: When straight parking is selected as the type of minimum risk strategy, the minimum risk strategy allowable space is set by using the longitudinal allowable distance and the lateral allowable distance; predicting a stoppable distance, a collision risk distance, and a lane departure distance, and determining the shortest distance among the predicted stoppable distance, the predicted collision risk distance, and the predicted lane departure distance as the longitudinal allowable distance; The smaller value between a value calculated by the adjacent lane width W-the width Wv of another vehicle traveling in the adjacent lane-0.75 m and the maximum intrusion allowable range is determined as the lateral allowable distance.

6. The vehicle according to claim 5, wherein: The processor is further configured to: A situation where the longitudinal permissible distance is shorter than the parking distance is predicted as a dangerous behavior.

7. The vehicle according to claim 6, wherein: When a dangerous behavior is predicted, the processor is further configured to: predicting at least one driving trajectory based on at least one braking force deviation; When the predicted driving trajectory based on the braking force deviation shows that the vehicle is parked within the minimum risk strategy allowed space, a braking force deviation to be applied to the vehicle is determined from the at least one braking force deviation.

8. The vehicle according to claim 7, wherein: The processor is further configured to: Update the minimum risk strategy allowable space when the vehicle turns due to asymmetric braking; The braking force bias is re-determined based on the updated minimum risk strategy allowance.

9. The vehicle according to claim 7, wherein: The processor is further configured to: Based on the driving trajectory information caused by the braking force deviation obtained through simulation or actual testing, the braking force deviation to be applied to the vehicle is determined by artificial intelligence using reinforcement learning.

10. The vehicle according to claim 9, wherein: The processor is further configured to: Information about the vehicle's speed, a preset minimum risk strategy allowance and information about the curvature of the current lane of travel are provided as inputs for the artificial intelligence to make the determination.

11. A method of operating a vehicle, comprising: Monitoring vehicle status information and / or surrounding environment information; Determining whether autonomous driving is possible based on vehicle state information and / or surrounding environment information; When autonomous driving is not possible, implement the minimum risk strategy; Predicting risky behavior, The executing of the minimum risk strategy includes executing the minimum risk strategy by additionally utilizing asymmetric braking to apply a braking force bias between left and right wheels of the vehicle when a dangerous behavior is predicted.

12. The method of operating a vehicle according to claim 11, wherein: Predictors of risky behaviors include: The situation where straight-line parking is selected as the type of the minimum risk strategy and the vehicle is traveling on a curved road is predicted as a dangerous behavior.

13. The method of operating a vehicle according to claim 12, wherein: Predictors of risky behaviors include: A situation in which straight-line parking is selected as the type of minimum risk strategy because lateral control is not partially or completely performed and the vehicle is traveling on a curved road is predicted as a dangerous behavior.

14. The method of operating a vehicle according to claim 12, wherein: Predictors of risky behaviors include: A lane departure, guardrail collision, or collision with another vehicle is predicted as a dangerous behavior.

15. The method of operating a vehicle according to claim 11, wherein: Determine the type of strategy with the least risk is to choose a straight line parking, Executing the minimum risk strategy further includes setting the minimum risk strategy allowable space, Setting the allowable space includes: Predict stopping distance, collision risk distance, and lane departure distance; Determine the shortest distance among the predicted stoppable distance, the predicted collision risk distance, and the predicted lane departure distance as the longitudinal allowable distance; The smaller value between the value calculated by the adjacent lane width W minus the width Wv of another vehicle traveling in the adjacent lane minus 0.75 m and the maximum intrusion allowable range is determined as the lateral allowable distance; The minimum risk strategy allowable space is set based on the longitudinal allowable distance and the lateral allowable distance.

16. The method of operating a vehicle according to claim 15, wherein: Predicting risky behavior further includes: A situation where the longitudinal permissible distance is shorter than the parking distance is predicted as a dangerous behavior.

17. The method of operating a vehicle according to claim 16, wherein: When risky behavior is predicted, minimal risk strategies can be implemented by additionally utilizing asymmetric braking including: predicting at least one driving trajectory based on at least one braking force deviation; When the predicted driving trajectory based on the braking force deviation shows that the vehicle is parked within the minimum risk strategy allowed space, a braking force deviation to be applied to the vehicle is determined from the at least one braking force deviation.

18. The method of operating a vehicle according to claim 17, wherein: When a dangerous behavior is predicted, executing a minimum risk strategy by additionally utilizing asymmetric braking further includes: Update the minimum risk strategy allowable space when the vehicle turns due to asymmetric braking; The braking force bias is re-determined based on the updated minimum risk strategy allowance.

19. The method of operating a vehicle according to claim 17, wherein: Determining the braking force bias to be applied to the vehicle includes: Based on the driving trajectory information caused by the braking force deviation obtained through simulation or actual testing, the braking force deviation to be applied to the vehicle is determined by artificial intelligence using reinforcement learning.

20. The method of operating a vehicle according to claim 19, wherein: Determining a braking force bias to be applied to the vehicle further includes: Information about the vehicle's speed, a preset minimum risk strategy allowance and information about the curvature of the current lane of travel are provided as inputs for the artificial intelligence to make the determination.

Images