Taking-over control method and device, vehicle and computer readable storage medium

Abstract

The embodiment of the invention provides a takeover control method and device, a vehicle and a computer readable storage medium, and the method comprises the steps: detecting the current state of a driver, and recognizing the takeover risk level of the vehicle according to the current state; the applied torque on a steering wheel of the vehicle, brake switch signal information and the opening degree of an accelerator pedal are collected, and the takeover parameters of the vehicle are determined according to at least one of the applied torque, the brake switch signal information and the opening degree of the accelerator pedal; and based on the takeover risk level and the takeover parameter, generating a takeover action of the vehicle, and controlling the vehicle to execute the takeover action. According to the method, the current state of the driver can be detected when the vehicle is in automatic driving, whether the current state of the driver is abnormal or not is recognized in time, and different takeover risk levels and parameter conditions are set according to different abnormal current states; therefore, whether the driver has misoperation or not and whether the driver needs to be restrained from taking over the vehicle or not are judged according to actual taking-over parameters of the driver, and safety of the vehicle, the driver and passengers is ensured.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a takeover control method, device, vehicle, and computer-readable storage medium. Background Technology

[0002] In related technologies, in vehicles with autonomous driving systems, if the user performs an action related to taking over the vehicle while it is in autonomous driving mode, the vehicle will return driving control to the user. However, the user may mistakenly take over, and allowing the user to take over the vehicle in such cases would pose a certain danger. Based on this, some studies analyze the vehicle's human-machine interaction data using predictive models to obtain a score of driver takeover in various traffic scenarios to determine whether or not to allow the driver to take over the vehicle.

[0003] However, in related technologies, traditional autonomous driving takeover methods or predictive models are based on vehicle and environmental scenarios, neglecting the important influence of the driver as the takeover operator during the takeover process. This can easily lead to driving risks due to driver errors during autonomous driving takeover, failing to guarantee the safety of the vehicle and its occupants, and urgently needs to be addressed. Summary of the Invention

[0004] This application provides a takeover control method, device, vehicle, and computer-readable storage medium, aiming to improve the problem that traditional autonomous driving takeover methods or prediction models in related technologies are based on vehicle and environmental scenarios, ignoring the important influence of the driver as the takeover person during the takeover process. This can easily lead to driving risks due to driver takeover errors during the autonomous driving takeover process, and cannot guarantee the safety of the vehicle and passengers.

[0005] A first aspect of this application provides a vehicle takeover control method, comprising the following steps: detecting the current state of the driver to identify the takeover risk level of the vehicle based on the current state; acquiring applied torque on the steering wheel, brake switch signal information, and accelerator pedal opening of the vehicle, and determining takeover parameters of the vehicle based on at least one of the applied torque, the brake switch signal information, and the accelerator pedal opening; generating a takeover action of the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action.

[0006] Through the above technical means, the embodiments of this application can generate different vehicle takeover actions for different driver takeover behaviors under different vehicle takeover risk levels and different takeover parameters, thereby controlling the vehicle to perform different takeover actions, such as allowing the driver to take over the vehicle or refusing the driver to take over the vehicle, so as to effectively maintain the driving safety of the vehicle in both normal and abnormal takeover situations.

[0007] Optionally, in one embodiment of this application, detecting the driver's current state to identify the vehicle's takeover risk level based on the current state includes: detecting the driver's facial state, seatbelt state, seat back angle, and gravity sensor information to determine the current state based on the facial state, seatbelt state, seat back angle, and gravity sensor information; determining the takeover risk level as Level 1 when the facial state is normally identifiable and not fatigued, the seatbelt is fastened, the seat back angle is greater than or equal to a target angle, and the gravity sensor information is greater than or equal to a gravity threshold; determining the takeover risk level as Level 2 when the facial state is abnormally identifiable or fatigued, the seatbelt is not fastened, or the gravity sensor information is less than the gravity threshold; and determining the takeover risk level as Level 3 when at least two of the following conditions are met: the facial state is abnormally identifiable, the facial state is fatigued, the seatbelt is not fastened, and the gravity sensor information is less than the gravity threshold.

[0008] Through the above technical means, the embodiments of this application can combine multi-dimensional detection of the driver's facial state, seat belt status, seat back angle, seat gravity sensor information, etc., to classify the risk level of driver takeover. It can remind the driver to take over in advance in the case of level 1, 2, and 3 risks to avoid being unable to react in time in abnormal situations. At the same time, it can determine whether the driver's takeover operation under the risk level is a mistake. Only after confirming that it is not a mistake can the driver take over. It can accurately determine the degree of risk and avoid takeover failure caused by mistake, thus taking into account both driving safety and the accuracy of takeover operation.

[0009] Optionally, in one embodiment of this application, generating the takeover action of the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: determining the takeover action as a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode when the takeover risk level is Level 1, and the duration of the applied torque being greater than or equal to the first target torque is greater than or equal to the first target duration, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening.

[0010] Through the above technical means, the embodiments of this application can verify the driver's ability to take over the vehicle under any of the following conditions when there is a level 1 risk of takeover: a relatively slight steering wheel torque and duration, vehicle parking, throttle opening, etc. This can not only ensure that the driver truly has the ability to control the vehicle by using clear indicators, eliminating formal takeover and erroneous takeover, but also ensure the safety and effectiveness of the vehicle when switching from autonomous driving to manual takeover.

[0011] Optionally, in one embodiment of this application, generating a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: when the takeover risk level is level two, and the duration of the applied torque being greater than the second target torque is greater than or equal to the second target duration, or the brake switch signal information is that the brake switch information is set and the duration of the brake pedal travel being greater than the target travel is greater than or equal to the third target duration, or the duration of the accelerator pedal opening being greater than the second target opening is greater than or equal to the fourth target duration, determining that the takeover action is a level two takeover action that allows the driver to take over the vehicle currently in autonomous driving mode, and performs the takeover risk level detection and / or the takeover parameter detection according to the target time interval within the target time; wherein, the second target duration is greater than the first target duration, and the second target opening is greater than the first target opening.

[0012] Through the above technical means, the embodiments of this application can verify the driver's ability to take over when the vehicle's takeover risk level is level two. This can be achieved by using any of the following conditions: the standard steering wheel torque and duration, the vehicle's parking time and duration, and the standard throttle opening and duration. This strictly requires the driver's driving operation and control abilities to prevent the driver from easily taking over and causing danger.

[0013] Optionally, in one embodiment of this application, generating the takeover action of the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: when the takeover risk level is level three, and the applied torque is zero, or the brake switch signal information is that the brake switch information is not set, or the accelerator pedal opening is zero, determining that the takeover action is a level three takeover action that prohibits the driver from taking over the vehicle currently in autonomous driving mode, and controlling the vehicle to initiate a parking action and a warning action.

[0014] Through the above technical means, the embodiments of this application can promptly determine that the driver lacks effective operating ability when the vehicle's takeover risk level is three, based on any condition such as no torque applied to the steering wheel, the brake not being engaged, or the accelerator opening being zero. At this time, the driver is prohibited from taking over the vehicle, and the autonomous driving parking and warning are triggered. This can not only avoid the risk of vehicle loss of control caused by the driver forcibly taking over due to inability to control, but also control the risk in the shortest possible time through proactive measures such as quickly pulling over to the side of the road and illuminating the hazard lights, while also alerting surrounding vehicles and maximizing the safety of driving, riding, and road traffic.

[0015] A second aspect of this application provides a vehicle takeover control device, comprising: a detection module for detecting the driver's current state to identify the vehicle's takeover risk level based on the current state; a data acquisition module for acquiring applied torque on the vehicle's steering wheel, brake switch signal information, and accelerator pedal opening, and determining the vehicle's takeover parameters based on at least one of the applied torque, the brake switch signal information, and the accelerator pedal opening; and a control module for generating a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action.

[0016] Through the above technical means, the embodiments of this application can generate different vehicle takeover actions for different driver takeover behaviors under different vehicle takeover risk levels and different takeover parameters, thereby controlling the vehicle to perform different takeover actions, such as allowing the driver to take over the vehicle or refusing the driver to take over the vehicle, to ensure that the driver can effectively maintain the driving safety of the vehicle in both normal and abnormal takeover situations.

[0017] Optionally, in one embodiment of this application, the detection module includes: a detection unit, configured to detect the driver's facial state, seat belt state, and seat back angle and gravity sensing information, to determine the current state based on the facial state, seat belt state, and seat back angle and gravity sensing information; a first determining unit, configured to determine the takeover risk level as Level 1 when the facial state is in a normally recognizable state and is not fatigued, the seat belt is fastened, the seat back angle is greater than or equal to a target angle, and the gravity sensing information is greater than or equal to a gravity threshold; a second determining unit, configured to determine the takeover risk level as Level 2 when the facial state is in an abnormally recognizable state or fatigued state, or the seat belt is not fastened, or the gravity sensing information is less than the gravity threshold; and a third determining unit, configured to determine the takeover risk level as Level 3 when at least two of the following conditions are met: the facial state is in an abnormally recognizable state, the facial state is fatigued, the seat belt is not fastened, and the gravity sensing information is less than the gravity threshold.

[0018] Through the above technical means, the embodiments of this application can combine multi-dimensional detection of the driver's facial state, seat belt status, seat back angle, seat gravity sensor information, etc., to classify the risk level of driver takeover. It can remind the driver to take over in advance in the case of level 1, 2, and 3 risks to avoid being unable to react in time in abnormal situations. At the same time, it can determine whether the driver's takeover operation under the risk level is a mistake. Only after confirming that it is not a mistake can the takeover be allowed. It can accurately match the risk level and avoid takeover failure caused by mistake, thus taking into account both driving safety and the accuracy of takeover operation.

[0019] Optionally, in one embodiment of this application, the control module includes: a fourth determining unit, configured to determine that the takeover action is a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode when the takeover risk level is Level 1, and the duration of the applied torque being greater than the first target torque is greater than or equal to the first target duration, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening.

[0020] Through the above technical means, the embodiments of this application can verify the driver's ability to take over the vehicle under any of the following conditions when there is a level 1 risk of takeover: a relatively slight steering wheel torque and duration, vehicle parking, throttle opening, etc. This can not only ensure that the driver truly has the ability to control the vehicle by using clear indicators, eliminating formal takeover and erroneous takeover, but also ensure the safety and effectiveness of the vehicle when switching from autonomous driving to manual takeover.

[0021] Optionally, in one embodiment of this application, the control module includes: a fifth determining unit, configured to determine that the takeover action is a level two takeover action that allows the driver to take over the vehicle currently in autonomous driving mode and performs the takeover risk level detection and / or the takeover parameter detection according to the target time interval within a target time, when the takeover risk level is level two, and the duration of the applied torque being greater than the second target torque is greater than or equal to the second target duration, or the duration of the brake switch signal information being brake switch information set and the duration of the brake pedal travel being greater than the target travel being greater than or equal to the third target duration, or the duration of the accelerator pedal opening being greater than the second target opening being greater than or equal to the fourth target duration; wherein the second target duration is greater than the first target duration, and the second target opening is greater than the first target opening.

[0022] Through the above technical means, the embodiments of this application can verify the driver's ability to take over when the vehicle's takeover risk level is level two. This can be achieved by using any of the following conditions: the standard steering wheel torque and duration, the vehicle's parking time and duration, and the standard throttle opening and duration. This strictly requires the driver's driving operation and control abilities to prevent the driver from easily taking over and causing danger.

[0023] Optionally, in one embodiment of this application, the control module includes: a sixth determining unit, configured to determine that the takeover action is a level three takeover action that prohibits the driver from taking over the vehicle currently in autonomous driving mode and controls the vehicle to initiate a parking action and a warning action when the takeover risk level is level three, the applied torque is zero, the brake switch signal information is that the brake switch information is not set, or the accelerator pedal opening is zero.

[0024] Through the above technical means, the embodiments of this application can promptly determine that the driver lacks effective operating ability when the vehicle's takeover risk level is three, based on any condition such as no torque applied to the steering wheel, the brake not being engaged, or the accelerator opening being zero. At this time, the driver is prohibited from taking over the vehicle, and the automatic driving system is triggered to stop and issue a warning. This not only avoids the risk of vehicle loss of control due to the driver's inability to control the vehicle but forcibly taking over, but also controls the risk in the shortest possible time through proactive measures such as quickly pulling over to the side of the road and activating the hazard lights, while also alerting surrounding vehicles and maximizing the safety of driving, riding, and road traffic. Attached Figure Description

[0025] Figure 1 This is a flowchart of a vehicle takeover control method provided according to an embodiment of this application; Figure 2 This is a flowchart illustrating the detection process of driver takeover according to one embodiment of this application; Figure 3This is a schematic diagram illustrating different takeover strategies based on different takeover behaviors of the driver, according to one embodiment of this application. Figure 4 This is a flowchart of a vehicle takeover control method based on user state detection according to an embodiment of this application; Figure 5 This is a structural diagram of the vehicle takeover control device provided in the embodiments of this application; Figure 6 This is a structural diagram of the vehicle provided in the embodiments of this application. Detailed Implementation

[0026] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0027] Before explaining the vehicle takeover control method in the embodiments of this application, the vehicles to which the vehicle takeover control method in the embodiments of this application is applicable will be explained first.

[0028] The vehicle takeover control method in this application embodiment is applicable to electric vehicles and internal combustion engine vehicles equipped with autonomous driving systems.

[0029] The autonomous driving system includes, but is not limited to, a lane line detection device, a forward target detection device, an adaptive cruise control module, an autonomous driving system control device, a system status display device, a DMS sensor, a driver's electric seat, and braking, steering, and drive systems.

[0030] Among them, the surrounding lane line detection device can use sensors such as smart cameras and lidar to detect the lane lines of the vehicle's driving lane and the surrounding lanes.

[0031] The forward target detection device can use cameras, millimeter-wave radar, and lidar to detect the position, status, color, and obstructing vehicles of traffic lights in left-turn intersections.

[0032] Drive, steering, and braking control devices can control the vehicle's acceleration, deceleration, and lateral control, generating corresponding commands to control the steering and braking systems, thus enabling comfortable control of autonomous driving vehicles.

[0033] The system status display device can inform the driver of their behavioral status through the driver interface, such as distraction, drowsiness, fatigue, driver absence, and seat belt status.

[0034] The adaptive cruise control module is used to automatically adjust the vehicle speed and following distance to assist driving.

[0035] The autonomous driving system control unit is used to coordinate and control all autonomous driving-related components of the vehicle to achieve autonomous driving functions.

[0036] DMS sensors are used to monitor information such as the driver's facial expression and attention to ensure driving safety.

[0037] This application provides a vehicle takeover control method, comprising: detecting the driver's current state to identify the vehicle's takeover risk level based on the current state; collecting the applied torque on the vehicle's steering wheel, brake switch signal information, and accelerator pedal opening, and determining the vehicle's takeover parameters based on the applied torque, brake switch signal information, and accelerator pedal opening; generating the vehicle's takeover action based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action.

[0038] This application embodiment can generate different vehicle takeover actions for different driver takeover behaviors under different vehicle takeover risk levels and takeover parameters, thereby controlling the vehicle to perform different takeover actions, such as allowing the driver to take over the vehicle or refusing the driver to take over the vehicle, to ensure that the driving safety of the vehicle can be effectively maintained in both normal and abnormal takeover by the driver.

[0039] Example 1 This application provides a vehicle takeover control method, please refer to... Figure 1 This includes the following steps: S110: Detect the driver's current state to identify the vehicle takeover risk level based on the current state; S120: Collects the applied torque on the vehicle's steering wheel, brake switch signal information, and accelerator pedal opening, and determines the vehicle's control parameters based on at least one of the applied torque, brake switch signal information, and accelerator pedal opening. S130: Based on the takeover risk level and takeover parameters, generate the takeover action for the vehicle and control the vehicle to execute the takeover action.

[0040] In real-world applications, autonomous driving, as a driver assistance function, requires the driver to take over the vehicle promptly when dangerous factors are detected around the vehicle, the vehicle is no longer suitable for autonomous driving, or the prescribed time has been reached.

[0041] However, in some scenarios, the driver's condition is not good when taking over the vehicle, or even unable to take over the vehicle, or may mistakenly take over the vehicle. In such scenarios, allowing the driver to take over the vehicle or directly disengaging from autonomous driving mode can easily lead to danger.

[0042] Based on this, embodiments of this application can determine the current takeover risk level of the vehicle by detecting the driver's current state. The driver's current state here includes, but is not limited to, the driver's physiological state, attention state, seatbelt status, and positional status. For example, whether the driver is distracted, drowsy, or fatigued; whether the seatbelt is properly fastened; and whether the driver is in a normal driving position. This clearly confirms the current takeover risk level of the vehicle, that is, the degree of risk when the driver takes over the vehicle in autonomous driving mode. For example, is there any risk in the driver taking over the vehicle, and what is the level of risk?

[0043] Furthermore, this application embodiment can detect vehicle takeover parameters, which can also be understood here as parameters of the driver's takeover behavior. That is, this application embodiment can detect parameters corresponding to some takeover behaviors made by the driver when actively taking over the vehicle or passively responding to the takeover request issued by the vehicle. For example, the magnitude of the applied torque on the steering wheel (the applied torque here refers to the steering hand torque, that is, the force applied by the driver to the steering wheel through the hands to control the vehicle's steering direction (such as turning left or right) or maintain the current driving direction), or brake switch signal information (information on the driver's control of the vehicle's start and stop), or accelerator pedal opening (the force or control of the driver on the accelerator pedal), etc. Through this information, it can be determined whether the driver has mistakenly taken over and whether he has the ability to take over the vehicle and maintain normal driving.

[0044] Therefore, the embodiments of this application can generate different vehicle takeover actions for different driver takeover behaviors (the actions of the driver when taking over) under different vehicle takeover risk levels (lower takeover risk = level 1; medium takeover risk = level 2; higher takeover risk = level 3) and different takeover parameters, thereby controlling the vehicle to perform different takeover actions, that is, allowing the driver to take over the vehicle, refusing the driver to take over the vehicle, and making other control actions, so as to effectively maintain the driving safety of the vehicle in both normal and abnormal takeover by the driver.

[0045] Step S110 includes detecting the driver's facial state, seat belt state, seat back angle, and gravity sensor information to determine the current state based on the facial state, seat belt state, seat back angle, and gravity sensor information; when the facial state is normally recognizable and the driver is not fatigued, the seat belt is fastened, the seat back angle is greater than or equal to the target angle, and the gravity sensor information is greater than or equal to the gravity threshold, the takeover risk level is determined to be Level 1; when the facial state is abnormally recognizable or the driver is fatigued, or the seat belt is not fastened, or the gravity sensor information is less than the gravity threshold, the takeover risk level is determined to be Level 2; when at least two of the following conditions are met: abnormally recognizable facial state, fatigued facial state, not fastened seat belt, or gravity sensor information less than the gravity threshold, the takeover risk level is determined to be Level 3.

[0046] In some embodiments, this application can detect the driver's facial state, seat belt state, and seat back angle and gravity sensor information to determine the driver's current state based on the driver's facial state, seat belt state, and seat back angle and gravity sensor information, thereby determining the vehicle's takeover risk level based on the driver's current state.

[0047] Among them, detecting the driver's facial state can clarify the driver's physiological and attentional state, such as concentration, distraction, fatigue, drowsiness, etc. Distraction will cause the driver's attention to deviate from the driving task, while drowsiness and fatigue reflect the driver's current physiological function decline, which will directly affect the driver's driving focus and reaction ability.

[0048] Checking the driver's seatbelt status means checking whether the driver is wearing the seatbelt correctly. It can also check whether the driver's driving behavior is compliant and effective, thus ensuring that even if the driver takes over the vehicle, the seatbelt is still properly fastened, thus protecting the driver's life.

[0049] By detecting the seat back angle and gravity sensor information, it is possible to confirm whether the driver is in the driver's seat, whether the driver is in a normal driving position / position, and whether the driver is maintaining a normal driving posture.

[0050] Specifically, when the driver's face is recognizable and not fatigued, the vehicle's seatbelt is fastened, the seat back angle is less than the target angle, and the gravity sensor information is greater than or equal to the gravity threshold, the vehicle's takeover risk level is determined to be no risk level.

[0051] When the driver's face is in a normally recognizable state (can be normally recognized by the relevant sensors in the vehicle) and is not fatigued, the seat belt is fastened, the seat back angle is less than the target angle, and the gravity sensor information is greater than or equal to the gravity threshold, the takeover risk level is determined to be no risk level.

[0052] In this context, "not fatigued" can be understood as the driver's facial state meeting the target attention condition, while "fatigued" can be understood as the driver's facial state not meeting the target attention condition.

[0053] Furthermore, the target attention condition here can be understood as a set of rules that the vehicle can use sensors (such as a DMS system) to determine whether the driver's attention is in a safe and focused driving state. For example, whether the eyes are focused on the road or ahead, whether the blinking frequency is within the normal range, whether there are excessive blinking characteristics of excessive fatigue or distraction, whether the head posture is at an angle conducive to observing road conditions, etc. When all these aspects meet the standard requirements for safe and focused driving, the target attention condition is met, and the driver is in a non-fatigued state. If any aspect is not met, it can be understood as not meeting the target attention condition, and the driver is in a fatigued state.

[0054] The target angle here refers to the lower limit of the seat back angle preset to ensure driver safety (and ease of operation). When the backrest angle is less than this value, it means that the driver's posture supports safe driving. For example, when the seat angle is greater than or equal to 150°, it is at an angle close to lying flat, indicating that the driver is likely lying down or leaning back, and the driving posture is unsafe and improper. When the seat angle is less than 150°, it is at a normal sitting posture where the driver can see the situation in front of the vehicle and can drive normally.

[0055] The gravity threshold here refers to the critical value at which the seat gravity sensor determines that the driver is in the driving position and in a normal posture. When the gravity sensing information (such as the driver's pressure on the seat) is greater than or equal to this threshold, it indicates that the driver is sitting in the driver's seat and the weight or pressure is consistent with the state of normal driving, rather than an abnormal situation such as partial contact with or removal from the seat.

[0056] It should be noted that the specific target attention conditions, target angle, and gravity threshold can be set or adjusted by those skilled in the art according to the actual situation. The embodiments in this application are only illustrative and do not impose any specific limitations.

[0057] When the face is in a normally recognizable state and not fatigued (meeting the target attention condition), the seat belt is fastened, and the gravity sensor information is greater than or equal to the gravity threshold, but the seat back angle is greater than or equal to the target angle, the takeover risk level is determined to be Level 1.

[0058] Furthermore, when the facial recognition status is abnormal, or the facial status is fatigued (does not meet certain attention conditions), or the seat belt status is not fastened, or the seat gravity sensor information is less than the gravity threshold, that is, when any of these four conditions are met, the takeover risk level is determined to be level two. Furthermore, the takeover risk level can be determined to be level three when at least two of the following conditions are met: the face is in an abnormal recognition state, the face is in a fatigued state (not meeting certain attention conditions), the seat belt is not fastened, and the seat's gravity sensor information is less than the gravity threshold.

[0059] For example, Table 1 is a takeover risk level judgment table according to one embodiment of this application, which can be represented as follows, but is not limited to:

[0060] Furthermore, when a vehicle detects that the user is at risk of being taken over at levels one, two, or three, the driver should be promptly reminded to take over the vehicle. This avoids having to remind the driver to take over the vehicle only when an abnormal situation occurs, as the driver may not be able to respond to the takeover request in a timely manner, thus ensuring driving safety.

[0061] Furthermore, when the vehicle detects that the driver is at risk of being taken over at levels one, two, or three, but still requests to take over the vehicle, the embodiments of this application can determine whether the driver is operating the vehicle incorrectly. Only if it is confirmed that the driver is not operating the vehicle incorrectly will the driver be allowed to take over the vehicle, which can effectively protect the driver's safety during abnormal takeover (mistaken takeover).

[0062] This application embodiment can combine multi-dimensional detection of driver facial status, seat belt status, seat back angle, seat gravity sensor information, etc., to classify the risk level of vehicle autonomous driving takeover. It can remind the driver to take over in advance in the case of level 1, 2, and 3 risks to avoid being unable to react in time in abnormal situations. It also helps to determine whether the driver's takeover operation under the risk level is a mistake. Only after confirming that it is not a mistake can the driver take over. It can accurately match the risk level and avoid takeover failure caused by mistake, thus balancing driving safety and the accuracy of takeover operation.

[0063] Step S130 includes determining that the takeover action is a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode when the takeover risk level is Level 1, and the duration of the applied torque being greater than or equal to the first target torque, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening.

[0064] It is understandable that the vehicle in autonomous driving mode and the vehicle used to determine the takeover risk level and detect takeover parameters are the same vehicle. It is only when the vehicle is in autonomous driving mode that the takeover risk level of the vehicle is determined and the takeover parameters of the vehicle are detected.

[0065] In actual implementation, the embodiments of this application can combine the takeover risk level and takeover parameters to determine different takeover actions for the vehicle. That is, the embodiments of this application can, under different vehicle takeover risk levels, use different standards to determine whether the driver has the ability to take over the vehicle normally, and whether the driver can take over the vehicle, based on the takeover parameters reflected by the driver's takeover behavior.

[0066] Figure 2 This is a flowchart illustrating the detection process of driver takeover according to one embodiment of this application. Figure 3 This is a schematic diagram illustrating different takeover strategies based on different driver takeover behaviors, according to one embodiment of this application. Figure 2 and Figure 3 As shown, when the current takeover risk level of the vehicle is detected to be Level 1, Level 2, and Level 3 based on the driver's current state, this embodiment of the application can determine the corresponding Level 1 takeover strategy, Level 2 takeover strategy, and Level 3 takeover strategy (corresponding to Level 1 takeover action, Level 2 takeover action, and Level 3 takeover action) by combining the takeover parameters corresponding to the driver's takeover behavior.

[0067] Specifically, when the vehicle takeover risk level is Level 1, it indicates that the danger of the driver taking over the vehicle is relatively low. Therefore, in this application embodiment, when determining the vehicle takeover action by combining the vehicle takeover risk level and takeover parameters, the requirements that the takeover parameters should meet can be set. That is, the driver can be judged whether he / she meets the conditions / standards for taking over the vehicle based on whether the takeover parameters meet certain conditions, and then the vehicle takeover action can be determined, such as the vehicle refusing the driver's takeover, allowing the driver to take over, and other operations.

[0068] In this embodiment of the application, when the vehicle's takeover risk level is Level 1, the conditions that the takeover parameters should meet are mainly based on the judgment that the driver's behavior is normal, there is no takeover risk, and the driver can control the vehicle's safety well. Therefore, the requirements for the takeover parameters should be comfortable and reasonable, and the process of switching from autonomous driving to human driving should be safer and smoother.

[0069] Based on this, the embodiments of this application can be set to the following when the vehicle takeover risk level is Level 1: (1) the applied torque on the steering wheel (the steering hand torque applied by the driver on the steering wheel) is greater than the first target torque, and the duration of the applied torque being greater than the first target torque is greater than or equal to the first target duration; (2) the brake switch signal information is set (i.e., the driver performs the action of pressing the brake pedal to brake, and the brake switch information is set accordingly); (3) the accelerator pedal opening (the change value of the accelerator pedal angle caused by the driver performing the action of pressing the accelerator pedal) is greater than the first target opening (the angle of the driver pressing the accelerator pedal is greater than the first target opening). When any one of these three conditions is met, the takeover action can be determined as a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode.

[0070] When the vehicle's takeover risk level is Level 1: as long as the driver can apply an actual torque on the steering wheel that is greater than the first target torque, and the duration of applying the torque is greater than or equal to the first target duration; or the driver performs a vehicle parking operation; or the accelerator pedal can be pressed to a degree greater than the first target opening, etc., if any one of these three conditions is met, the embodiment of this application can determine that the driver can take over the vehicle well, and the driver can be allowed to take over the vehicle currently in autonomous driving mode.

[0071] Here, the first target torque can be understood as the minimum torque threshold value that the vehicle is pre-set to determine whether the driver's steering wheel control force is effective when the vehicle's takeover risk level is level one. That is, when the vehicle's takeover risk level is level one, the driver needs to apply a torque of no less than this value on the steering wheel in order to demonstrate effective control of the steering wheel.

[0072] The first target duration here refers to the minimum critical value set in advance by the vehicle to determine whether the duration for which the driver applies effective torque (torque greater than the first target torque) meets the standard when the vehicle's takeover risk level is Level 1. In other words, when the vehicle's takeover risk level is Level 1, the driver must maintain the state of applying effective torque for no less than this duration in order to demonstrate the ability to continuously control the steering wheel.

[0073] The first target opening here refers to the minimum accelerator pedal depressing threshold preset by the vehicle, which is used to determine that the driver can effectively control the vehicle's power when the vehicle's takeover risk level is Level 1. In other words, when the vehicle's takeover risk level is Level 1, this opening must be reached to prove that the driver has the operational ability to actively adjust the vehicle's power.

[0074] It should be noted that the first target torque, the first target duration, and the first target opening can be set to 3 Nm, 200 ms, and 5%, respectively, in this embodiment of the application. In actual application, these can be set or adjusted by those skilled in the art according to the actual situation. This embodiment of the application is only for illustrative purposes and does not make any specific requirements.

[0075] The embodiments of this application can verify the driver's ability to take over the vehicle under any of the following conditions when there is a level 1 risk of takeover: a relatively slight steering wheel torque and duration, vehicle parking, throttle opening and duration. This can ensure that the driver has real vehicle control ability through clear indicators, eliminate formal takeover and false takeover, and ensure the safety and effectiveness of the vehicle when switching from autonomous driving to manual takeover.

[0076] Step S130 includes determining that the takeover action is a Level 2 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode, and performs takeover risk level detection and / or takeover parameter detection according to the target time interval within the target time, when the takeover risk level is Level 2 and the duration of the applied torque being greater than the second target torque is greater than or equal to the second target duration, or the duration of the brake switch signal information being brake switch information set and the duration of the brake pedal travel being greater than the target travel being greater than or equal to the third target duration, or the duration of the accelerator pedal opening being greater than the second target opening being greater than or equal to the fourth target duration; wherein, the second target duration is greater than the first target duration and the second target opening is greater than the first target opening.

[0077] In other embodiments, when the vehicle takeover risk level is level two, it is demonstrated that the danger of the driver taking over the vehicle is moderate. Therefore, when determining the takeover action of the vehicle in combination with the vehicle takeover parameters, the requirements that the takeover parameters should meet are more stringent than the requirements set when the vehicle takeover risk level is level one.

[0078] In this embodiment of the application, when the vehicle's takeover risk level is level two, the conditions for setting the takeover parameters are mainly based on deliberately avoiding the driver's unintentional mistakeover. Therefore, the designed parameter conditions should be sufficient to reflect the driver's subjective awareness, because a vehicle's takeover risk level of level two may lead the autonomous driving system to judge that the driver's behavior is risky. Taking over easily will be considered a mistake, and the vehicle cannot be controlled in time after taking over. At this time, switching to manual takeover is likely to lead to an accident. Therefore, the parameter conditions should be more stringent.

[0079] Based on this, the embodiments of this application can be set to the following when the vehicle takeover risk level is level two: (1) the applied torque on the steering wheel (the torque applied by the driver on the steering wheel) is greater than the second target torque, and the duration of the applied torque being greater than the second target torque should be greater than or equal to the second target duration; (2) the brake switch signal information is brake switch information set and the duration of the brake pedal travel (the change in brake pedal angle caused by the driver's brake pedal action) being greater than the target travel is greater than or equal to the third target duration; (3) when the accelerator pedal opening is greater than the second target opening (the angle of the driver pressing the accelerator pedal is greater than the second target opening) and the duration of the second target opening is greater than or equal to the fourth target duration, the takeover action is determined to be a level two takeover that allows the driver to take over the vehicle currently in autonomous driving mode.

[0080] When the vehicle's takeover risk level is Level 2, the driver can be deemed capable of taking over the vehicle if: 1) the driver can apply an actual torque on the steering wheel greater than the second target torque, and the duration of applying this torque is greater than or equal to the second target duration; 2) the driver can depress the brake pedal to stop the vehicle, and the brake pedal travel is greater than the target travel, and the duration of the brake pedal travel greater than the target travel is greater than or equal to the third target duration; or 3) the driver can depress the accelerator pedal to an opening greater than the second target opening, and the duration of the second target opening is greater than or equal to the fourth target duration.

[0081] Here, the second target torque can be understood as the minimum torque threshold value that the vehicle is pre-set to determine whether the driver's steering wheel control force is effective when the vehicle's takeover risk level is level two. That is, when the vehicle's takeover risk level is level two, the driver needs to apply a torque of no less than this value on the steering wheel in order to demonstrate effective control of the steering wheel.

[0082] The second target duration here refers to the minimum critical value set in advance by the vehicle to determine whether the duration for which the driver applies effective torque (torque greater than the second target torque) meets the standard when the vehicle's takeover risk level is level two. In other words, when the vehicle's takeover risk level is level two, the driver must maintain the state of applying effective torque (torque greater than the second target torque) for no less than this duration in order to demonstrate the ability to continuously control the steering wheel.

[0083] The target travel distance here refers to the minimum brake pedal depressing threshold preset by the vehicle, which is used to determine that the driver can effectively control the vehicle's power when the vehicle's takeover risk level is level two. In other words, when the vehicle's takeover risk level is level two, this travel distance must be reached to prove that the driver has the operational ability to actively stop the vehicle's power.

[0084] The third target duration here refers to the minimum critical value set in advance by the vehicle to determine whether the driver's effective braking force (the force corresponding to the brake pedal travel (angle) being greater than the target travel) is met when the vehicle's takeover risk level is level two. In other words, when the vehicle's takeover risk level is level two, the driver must maintain the state of applying effective braking force for no less than this duration in order to demonstrate the ability to continuously control the brake pedal.

[0085] The second target opening here refers to the minimum accelerator pedal depressing threshold preset by the vehicle when the vehicle's takeover risk level is level two, which is used to determine that the driver can effectively control the vehicle's power. In other words, when the vehicle's takeover risk level is level two, this opening must be reached to prove that the driver has the operational ability to actively adjust the vehicle's power.

[0086] The fourth target duration here can refer to the minimum critical value set by the vehicle in advance, which determines whether the duration of the driver applying effective throttle force (the force corresponding to the throttle pedal angle being greater than the second target opening) meets the standard when the vehicle's takeover risk level is level two. In other words, when the vehicle's takeover risk level is level two, the driver must maintain the state of applying effective throttle force for no less than this duration in order to demonstrate the ability to continuously control the throttle pedal.

[0087] It should be noted that the second target torque, second target duration, target stroke, third target duration, second target opening degree, and fourth target duration in this embodiment can be, but are not limited to, set to 8 Nm, 2 s, 20%, 2 s, 5%, and 2 s, respectively. In practical applications, the second target torque, second target duration, target stroke, third target duration, second target opening degree, and fourth target duration can all be set or adjusted by those skilled in the art according to the actual situation. This embodiment is only illustrative and does not impose specific limitations.

[0088] In addition, besides the first and second target durations, which both correspond to the duration of applied torque, the third target duration corresponding to the duration of brake pedal travel and the fourth target duration corresponding to the duration of accelerator pedal opening can be set the same as or different from the first or second target durations.

[0089] Furthermore, in order to ensure that drivers taking over under the Level 2 takeover risk level have the ability to take over for an extended period of time, embodiments of this application may also perform takeover risk level detection and / or takeover parameter detection at target time intervals within a target time period (takeover risk level detection or takeover parameter detection or both takeover risk level detection and takeover parameter detection may be performed).

[0090] Here, the target time can be understood as a pre-set time period for detecting the driver's status after the driver takes over the vehicle under the Level 2 takeover risk level, such as 5 minutes; the target time interval refers to the fixed time difference between two takeover risk level detections or takeover parameter detections, such as a detection every 1 minute, which is used to control the detection frequency.

[0091] Therefore, the embodiments of this application can detect whether the driver's state after taking over the vehicle meets the requirements of a takeover state (such as focused attention, no drowsiness or fatigue) for a certain duration and has continuous driving ability, indicating that the driver can continuously drive the vehicle. It should be noted that in practical applications, the target time and target time interval can be set or adjusted by those skilled in the art according to the actual situation. The embodiments of this application are only illustrative and do not make specific requirements.

[0092] This application embodiment can verify the driver's ability to take over when the vehicle's takeover risk level is level two. It can do so by using any of the following conditions: the standard steering wheel torque and duration, the vehicle's parking time and duration, and the standard throttle opening and duration. This strictly requires the driver's driving operation and control abilities to prevent the driver from easily taking over and causing danger.

[0093] Step S130 includes determining the takeover action as a Level 3 takeover action when the takeover risk level is Level 3, the applied torque is zero, or the brake switch signal information is that the brake switch information is not set, or the accelerator pedal opening is zero, prohibiting the driver from taking over the vehicle currently in autonomous driving mode, and controlling the vehicle to start parking and warning actions.

[0094] In some embodiments, when the vehicle's takeover risk level reaches level three, it indicates that the driver's existing driving and control abilities are extremely lacking, essentially rendering them incapable of taking over at all. Allowing the driver to take over the vehicle in this situation could pose a significant accident risk. For example, if the driver's face is turned towards the rear seats, meaning their eyes are not focused on the road ahead, and the steering wheel is suddenly subjected to a large intervention torque, directly disengaging the automatic driving system and handing over control to the driver would be risky. Because the driver is not looking ahead and is unaware of the risks, they cannot determine whether driving is safe, greatly increasing the probability of an accident. Therefore, in such cases, driver takeover should be discouraged. The automatic driving system should remain active, continuing to control the vehicle safely and reminding the driver to stay focused and ready to take over at any time, thus improving driving safety.

[0095] Based on this, when setting the takeover action when the vehicle takeover risk level is three, the embodiments of this application mainly focus on suppressing driver takeover and set certain parameter conditions accordingly.

[0096] When the risk level of vehicle takeover is determined to be Level 3, regardless of whether the driver takes over actively or passively, if any one of the following three conditions is met: the applied torque on the steering wheel is zero, or the brake switch signal is not set (i.e., no force is applied to the brake pedal and braking cannot be performed), or the accelerator pedal opening is zero (i.e., no force is applied to the accelerator pedal and accelerator pedal cannot be pressed), the driver is immediately prohibited from taking over the vehicle currently in autonomous driving mode, and the vehicle is controlled to initiate parking and warning actions. That is, the vehicle continues to be controlled by the autonomous driving system, and parking and warning actions are performed through the autonomous driving system. For example, the autonomous driving system will park the vehicle on the side of the road in the shortest possible time and activate the hazard lights.

[0097] This application embodiment can promptly determine that the driver lacks effective operational ability when the vehicle's takeover risk level is three, based on any condition such as no torque applied to the steering wheel, the brake not being engaged, or the accelerator opening being zero. At this time, the driver is prohibited from taking over the vehicle, and the autonomous driving system is triggered to stop and issue a warning. This not only avoids the risk of vehicle loss of control due to the driver's inability to control the vehicle but also enables proactive measures such as quickly pulling over to the side of the road and activating hazard lights through autonomous driving, controlling the risk in the shortest possible time, while also alerting surrounding vehicles, thus maximizing the safety of driving passengers and road traffic.

[0098] The following is a detailed explanation of the vehicle takeover control method in this application embodiment using a specific example.

[0099] Figure 4 This is a flowchart illustrating a vehicle takeover control method based on user state detection, according to one embodiment of this application. Figure 4 As shown: Step S410: Obtain information related to the driver's current state, including but not limited to the driver's facial expression, head position, seat belt status, driver's seat position, seat back angle, etc. Step S420: Determine the vehicle takeover risk level based on the information obtained in step S410; Step S430: Based on the vehicle's takeover risk level, formulate different takeover strategies, such as takeover conditions and warning actions and parking actions to prevent takeover. Step S440: The actuator controlling the vehicle responds to various takeover logics and determines whether the driver is allowed to take over the vehicle or not, based on the actual situation.

[0100] This application embodiment can detect the driver's current state when the vehicle is in autonomous driving mode, and promptly identify whether the driver's current state is abnormal, such as drowsiness, lack of concentration, or unfastening the seat belt. Different takeover risk levels are set for different abnormal current states, and parameter conditions are set specifically for different takeover risk levels. In order to determine whether the driver has made a mistake and whether he can take over the vehicle normally, based on the driver's actual takeover parameters, and take actions to inhibit the driver's takeover intervention when necessary, so as to ensure the safety of the vehicle and the passengers.

[0101] This application also provides a vehicle takeover control device 50, please refer to... Figure 5 The system includes: a detection module 510 for detecting the driver's current state to identify the vehicle's takeover risk level based on the current state; a data acquisition module 520 for acquiring the applied torque on the vehicle's steering wheel, brake switch signal information, and accelerator pedal opening, and determining the vehicle's takeover parameters based on at least one of the applied torque, brake switch signal information, and accelerator pedal opening; and a control module 530 for generating the vehicle's takeover action based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action.

[0102] The detection module 510 includes: a detection unit for detecting the driver's facial state, seat belt state, seat back angle, and gravity sensing information to determine the current state based on the facial state, seat belt state, seat back angle, and gravity sensing information; a first determination unit for determining the takeover risk level as Level 1 when the facial state is normally recognizable and the driver is not fatigued, the seat belt is fastened, the seat back angle is greater than or equal to a target angle, and the gravity sensing information is greater than or equal to a gravity threshold; a second determination unit for determining the takeover risk level as Level 2 when the facial state is abnormally recognizable or the driver is fatigued, the seat belt is not fastened, or the gravity sensing information is less than a gravity threshold; and a third determination unit for determining the takeover risk level as Level 3 when at least two of the following conditions are met: abnormally recognizable facial state, fatigued facial state, unfastened seat belt, and gravity sensing information less than a gravity threshold.

[0103] The control module 530 includes: a fourth determining unit, used to determine that the takeover action is a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode when the takeover risk level is Level 1 and the duration of the applied torque being greater than or equal to the first target torque, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening.

[0104] The control module 530 includes: a fifth determining unit, used to determine the takeover action as a level two takeover action that allows the driver to take over the vehicle currently in autonomous driving mode, and performs takeover risk level detection and / or takeover parameter detection according to the target time interval within the target time, when the takeover risk level is level two, and the duration of the applied torque being greater than the second target torque is greater than or equal to the second target duration, or the brake switch signal information is brake switch information set and the duration of the brake pedal travel being greater than the target travel being greater than or equal to the third target duration, or the duration of the accelerator pedal opening being greater than the second target opening being greater than or equal to the fourth target duration; wherein the second target duration is greater than the first target duration, and the second target opening is greater than the first target opening.

[0105] The control module 530 includes: a sixth determining unit, used to determine the takeover action as a level three takeover action that prohibits the driver from taking over the vehicle currently in autonomous driving mode and controls the vehicle to start a parking action and a warning action when the takeover risk level is level three, the applied torque is zero, or the brake switch signal information is that the brake switch information is not set, or the accelerator pedal opening is zero.

[0106] This application also provides a vehicle 60, please refer to... Figure 6 It includes a processor 610 and a memory 620, wherein the memory 610 is used to store computer programs; and the processor 620 is used to execute the programs stored in the memory 610 to implement the vehicle takeover control method described in any embodiment of this application.

[0107] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle takeover control method described in any embodiment of this application.

[0108] In this application, "multiple" refers to two or more.

[0109] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0110] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0111] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0112] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if the method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if the method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.

[0113] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for taking over control of a vehicle, characterized in that, Includes the following steps: Detect the driver's current state to identify the vehicle's takeover risk level based on the current state; The system collects the applied torque on the vehicle's steering wheel, brake switch signal information, and accelerator pedal opening, and determines the vehicle's control parameters based on at least one of the applied torque, brake switch signal information, and accelerator pedal opening. Based on the takeover risk level and the takeover parameters, a takeover action is generated for the vehicle, and the vehicle is controlled to execute the takeover action.

2. The method according to claim 1, characterized in that, The method of detecting the driver's current state to identify the vehicle's takeover risk level based on the current state includes: The driver's facial expression, seat belt status, seat back angle, and gravity sensor information are detected to determine the current state based on the facial expression, seat belt status, and seat back angle and gravity sensor information. When the face is in a normally recognizable state and is not fatigued, the seat belt is fastened, the seat back angle is greater than or equal to the target angle, and the gravity sensor information is greater than or equal to the gravity threshold, the takeover risk level is determined to be Level 1. If the facial recognition status is abnormal or the person is fatigued, or the seatbelt status is not fastened, or the gravity sensor information is less than the gravity threshold, the takeover risk level is determined to be level two. When at least two of the following conditions are met: the facial state is in the abnormal recognition state, the facial state is in the fatigue state, the seat belt state is in the unfastened state, and the gravity sensing information is less than the gravity threshold, the takeover risk level is determined to be level three.

3. The method according to claim 1, characterized in that, The process of generating a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: When the takeover risk level is Level 1, and the duration of the applied torque being greater than the first target torque is greater than or equal to the first target duration, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening, the takeover action is determined to be a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode.

4. The method according to claim 1, characterized in that, The process of generating a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: When the takeover risk level is level two, and the duration of the applied torque being greater than the second target torque is greater than or equal to the second target duration, or the brake switch signal information is brake switch information set and the duration of the brake pedal travel being greater than the target travel is greater than or equal to the third target duration, or the duration of the accelerator pedal opening being greater than the second target opening is greater than or equal to the fourth target duration, the takeover action is determined to be a level two takeover action that allows the driver to take over the vehicle currently in autonomous driving mode, and performs the takeover risk level detection and / or the takeover parameter detection according to the target time interval within the target time. Wherein, the duration of the second target is greater than the duration of the first target, and the opening degree of the second target is greater than the opening degree of the first target.

5. The method according to claim 1, characterized in that, The process of generating a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and controlling the vehicle to execute the takeover action, includes: When the takeover risk level is level three, and the applied torque is zero, or the brake switch signal information is that the brake switch information is not set, or the accelerator pedal opening is zero, the takeover action is determined to be a level three takeover action that prohibits the driver from taking over the vehicle currently in autonomous driving mode and controls the vehicle to initiate parking and warning actions.

6. A vehicle takeover control device, characterized in that, include: A detection module is used to detect the driver's current state in order to identify the takeover risk level of the vehicle based on the current state; The acquisition module is used to acquire the applied torque on the steering wheel of the vehicle, the brake switch signal information and the accelerator pedal opening, and to determine the vehicle's control parameters based on at least one of the applied torque, the brake switch signal information and the accelerator pedal opening. The control module is used to generate a takeover action for the vehicle based on the takeover risk level and the takeover parameters, and to control the vehicle to execute the takeover action.

7. The apparatus according to claim 6, characterized in that, The detection module includes: The detection unit is used to detect the driver's facial state, seat belt state, and seat back angle and gravity sensing information, so as to determine the current state based on the facial state, seat belt state, and seat back angle and gravity sensing information. The first determining unit is configured to determine the takeover risk level as Level 1 when the facial state is in a normally recognizable state and is not fatigued, the seat belt state is fastened, the seat back angle is greater than or equal to the target angle, and the gravity sensing information is greater than or equal to the gravity threshold. The second determining unit is used to determine the takeover risk level as level two when the facial state is in an abnormal recognition state or a fatigued state, or the seat belt state is in an unfastened state, or the gravity sensing information is less than the gravity threshold. The third determining unit is used to determine the takeover risk level as level three when at least two of the following conditions are met: the facial state is the abnormal recognition state, the facial state is the fatigue state, the seat belt state is the unfastened state, and the gravity sensing information is less than the gravity threshold.

8. The apparatus according to claim 6, characterized in that, The control module includes: The fourth determining unit is used to determine that the takeover action is a Level 1 takeover action that allows the driver to take over the vehicle currently in autonomous driving mode when the takeover risk level is Level 1, and the duration of the applied torque being greater than or equal to the first target torque is greater than or equal to the first target duration, or the brake switch signal information is brake switch information set, or the accelerator pedal opening is greater than the first target opening.

9. A vehicle, characterized in that, Including processor and memory, among which Memory, used to store computer programs; A processor for executing a program stored in memory to implement the method described in any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-5.

Images