Method, system and vehicle for safety takeover of an assisted driving system

Abstract

The present application relates to the technical field of automatic driving of automobiles, and discloses a safety takeover method and system of an auxiliary driving system and a vehicle, wherein the intelligent driving controller and the steering controller are independent devices, the intelligent driving controller sends an auxiliary driving exit instruction to other associated controllers that act on the auxiliary driving when it is determined that the conditions for exiting the auxiliary driving function are met without any fault of the intelligent driving controller, and the steering controller sends the auxiliary driving exit instruction to the intelligent driving controller when it is determined that the conditions for exiting the auxiliary driving function are met without any fault of the steering controller after detecting a fault of the intelligent driving controller, so as to realize the safety takeover of the vehicle, and the other device without any fault can still exit the auxiliary driving function when either the intelligent driving controller or the steering controller is faulty, thereby achieving the effect of double protection and ensuring that the driver can successfully take over the vehicle in any case.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving of automobiles, and in particular to a safety takeover method and system for an auxiliary driving system and a vehicle. BACKGROUND

[0002] With the continuous development of new generation information technologies such as artificial intelligence, 5G, and big data, intelligent networked automobile technology has a strong momentum of development, and its market share is becoming higher and higher. The main functions of an advanced driver assistance system (ADAS) include integrated adaptive cruise control (IACC), adaptive cruise control (ACC), lane keeping aid (LKA), autonomous emergency braking (AEB), auto parking assist (APA), lane departure warning (LDW), and blind spot detection (BSD). These functions improve the driving experience and make automobiles more intelligent.

[0003] However, in order to realize the above functions, the electronic and electrical system of the whole vehicle becomes more complex, and the probability of failure or malfunction of the electronic and electrical system becomes higher. In the IACC cruise process, a failure of the steering system (EPS) or a failure of the ADAS domain control (ADC) may cause the driver to be unable to take over the vehicle, thereby causing unacceptable safety risks. SUMMARY

[0004] Therefore, the present application provides a safety takeover method and system for an auxiliary driving system and a vehicle to solve the problem that in the IACC cruise process, a failure of the steering system or a failure of the ADAS domain control may cause the driver to be unable to take over the vehicle, thereby causing unacceptable safety risks.

[0005] In a first aspect, the application provides a safety takeover method of an assisted driving system, applied to an intelligent driving controller of a vehicle, the method comprising: in the absence of a fault of the intelligent driving controller, exiting an assisted driving function based on a hand torque and a duration of the hand torque applied to a steering wheel by a driver, and sending an assisted driving exit instruction to other associated controllers acting on the assisted driving, so as to cause the other associated controllers to exit the assisted driving function; or, in response to receiving an assisted driving exit instruction sent by a steering controller of the vehicle, exiting the assisted driving function, and sending the assisted driving exit instruction to the other associated controllers acting on the assisted driving, so as to cause the other associated controllers to exit the assisted driving function, the assisted driving exit instruction being an assisted driving exit instruction sent by the steering controller to the intelligent driving controller based on the hand torque and the duration of the hand torque applied to the steering wheel by the driver when the intelligent driving controller is detected to have a fault, but the steering controller is absent of a fault.

[0006] The safety takeover method of the assisted driving system provided by the application is characterized in that the intelligent driving controller and the steering controller are independent devices, and can both exit the assisted driving function based on the hand torque and the duration of the hand torque applied to the steering wheel by the driver in the absence of a fault of the intelligent driving controller, wherein the intelligent driving controller can send the assisted driving exit instruction to the other associated controllers acting on the assisted driving when it is determined that the conditions for exiting the assisted driving function are met, regardless of whether the steering controller has a fault or not, and the steering controller can send the assisted driving exit instruction to the intelligent driving controller when it is determined that the conditions for exiting the assisted driving function are met in the case that the intelligent driving controller is detected to have a fault, but the steering controller is absent of a fault, so as to realize the complete takeover of the driver, and the other device absent of a fault can still exit the assisted driving function when the intelligent driving controller or the steering controller has a fault, thereby achieving the effect of double protection and ensuring that the driver can successfully take over the vehicle in any case.

[0007] In an optional embodiment, the method further comprises that the hand torque and the duration of the hand torque applied to the steering wheel by the driver are the result of the driver operating the steering wheel to perform a safety takeover operation, and the driver operating the steering wheel to perform the safety takeover operation is an operation performed by the driver after the intelligent driving controller prompts the driver to perform the safety takeover of the vehicle in response to receiving the fault information sent by the steering controller or detecting that the intelligent driving controller has a fault.

[0008] The steering controller of the application can send the fault information to the intelligent driving controller when the steering controller has a fault, and the intelligent driving controller can prompt the driver to perform the safety takeover of the vehicle based on the fault information sent by the intelligent driving controller or detecting that the intelligent driving controller has a fault, so as to timely notify the driver to take over the vehicle when the vehicle has a fault.

[0009] In an optional implementation, the exiting the assisted driving function and sending the assisted driving exiting instruction to the other associated controllers for the assisted driving based on the hand torque and the duration of the hand torque applied by the driver to the steering wheel includes: determining whether the hand torque applied by the driver to the steering wheel is greater than a first hand torque threshold; if the hand torque applied by the driver to the steering wheel is greater than the first hand torque threshold, recording a first duration of the hand torque applied by the driver to the steering wheel being greater than the first hand torque threshold; and when the first duration is greater than a first time threshold, exiting the assisted driving function and sending the assisted driving exiting instruction to the other associated controllers for the assisted driving.

[0010] In an optional implementation, the intelligent driving controller integrates a threshold prediction model corresponding to different driving scenarios to different hand torque thresholds and time thresholds, and the method further includes: obtaining a driving scenario of the current vehicle; and inputting the driving scenario of the current vehicle into the threshold prediction model to obtain a first hand torque threshold and a first time threshold corresponding to the driving scenario of the current vehicle.

[0011] In a second aspect, the present application provides a safety takeover method of an assisted driving system, applied to a steering controller of a vehicle, and the method includes: in a case where the intelligent driving controller is detected to have a fault but the steering controller itself does not have a fault, exiting an assisted driving function based on a hand torque and a duration of the hand torque applied by a driver to a steering wheel, and sending an assisted driving exiting instruction to an intelligent driving controller of the vehicle, so that the intelligent driving controller exits the assisted driving function based on the assisted driving exiting instruction sent by the steering controller, and sends an assisted driving exiting instruction to other associated controllers for the assisted driving, so that the other associated controllers exit the assisted driving function.

[0012] In an optional implementation, the method further includes: detecting whether the steering controller itself has a fault, and when the steering controller itself is detected to have a fault, sending fault information to the intelligent driving controller, so that the intelligent driving controller prompts a driver to perform a safety takeover of the vehicle based on the fault information sent by the steering controller, and in a case where the intelligent driving controller itself does not have a fault, exiting the assisted driving function based on the hand torque and the duration of the hand torque applied by the driver to the steering wheel, and sending the assisted driving exiting instruction to the other associated controllers for the assisted driving, so that the other associated controllers exit the assisted driving function.

[0013] In an optional embodiment, the exiting the assisted driving function and sending the assisted driving exit instruction to the intelligent driving controller based on the hand torque of the driver acting on the steering wheel and the duration includes: determining whether the hand torque of the driver acting on the steering wheel is greater than a second hand torque threshold; if the hand torque of the driver acting on the steering wheel is greater than the second hand torque threshold, recording a second duration of the hand torque of the driver acting on the steering wheel being greater than the second hand torque threshold; and when the second duration is greater than a second time threshold, sending the assisted driving exit instruction to the intelligent driving controller, the second hand torque threshold being greater than a first hand torque threshold, the first hand torque threshold being a threshold used by the intelligent driving controller to determine whether to exit the assisted driving function based on the hand torque of the driver acting on the steering wheel.

[0014] The intelligent driving controller is used as a first layer protection mechanism, and the steering controller is used as a second layer protection mechanism, so that the problem that the driver cannot take over the vehicle through the steering wheel when the safety mechanism of the intelligent driving controller or the steering controller fails can be solved, a double protection effect is achieved, and it is ensured that the driver can successfully take over the vehicle in any case.

[0015] In an optional embodiment, the exiting the assisted driving function and sending the assisted driving exit instruction to the intelligent driving controller based on the hand torque of the driver acting on the steering wheel and the duration further includes: if it is determined that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, recording a third duration of the hand torque of the driver acting on the steering wheel being greater than the first hand torque threshold but less than the second hand torque threshold; and when the third duration is greater than a third time threshold, sending the assisted driving exit instruction to the intelligent driving controller.

[0016] The steering controller is designed to send the assisted driving exit instruction to the intelligent driving controller when the third duration is greater than the third time threshold, if it is determined that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, so that the risk that the driver cannot immediately take over the vehicle when the actual hand torque of the driver does not reach the second hand torque threshold is avoided.

[0017] In an optional embodiment, the steering controller integrates a threshold prediction model corresponding to different hand torque thresholds and time thresholds in different driving scenes, and the method further includes: obtaining a driving scene of the current vehicle; inputting the driving scene of the current vehicle into the threshold prediction model to obtain a second hand torque threshold, a second time threshold, a third hand torque threshold, and a third time threshold corresponding to the driving scene of the current vehicle.

[0018] The application determines the corresponding hand force threshold and time length threshold based on the driving scene of the current vehicle, and judges whether the hand force torque and the duration of the current driver acting on the steering wheel meet the conditions for the assistant driving function to exit based on the determined hand force threshold and time length threshold, which can improve the smoothness and safety of the driver's takeover.

[0019] In an optional embodiment, the threshold prediction model corresponding to different driving scenes and different hand force thresholds and time length thresholds is obtained by the following steps: based on different data acquisition methods, corresponding safety takeover data in all driving scenes is respectively acquired, wherein the safety takeover data in any driving scene includes safety takeover data acquired in different data acquisition methods; all safety takeover data in the driving scenes is respectively configured into the corresponding driving scene set according to the preset driving scene set corresponding to different hand force thresholds and time length thresholds, and the average hand force torque and the average duration corresponding to different data acquisition methods under different hand force thresholds and time length thresholds are calculated, wherein the different hand force thresholds and time length thresholds include the first hand force threshold and the first time length threshold, the second hand force threshold and the second time length threshold, and the third hand force threshold and the third time length threshold; the weights corresponding to different data acquisition methods in the preset current driving scene are weighted and summed with the average hand force torque and the average duration corresponding to different data acquisition methods under different hand force thresholds and time length thresholds to obtain the different hand force thresholds and time length thresholds corresponding to the current driving scene; all driving scenes are traversed, and the step of weighting and summing the weights corresponding to different data acquisition methods in the current driving scene with the average hand force torque and the average duration corresponding to different data acquisition methods under different hand force thresholds and time length thresholds is repeatedly executed to obtain the different hand force thresholds and time length thresholds corresponding to different driving scenes; the different hand force thresholds and time length thresholds corresponding to different driving scenes are input into the preset neural network model for training to obtain the threshold prediction model.

[0020] The application can allocate weights according to the characteristics of different driving scenes, enhance the adaptability of the system to different driving scenes and drivers, and reduce the safety risks caused by improper takeover.

[0021] In a third aspect, the present application provides a safety takeover system of an assisted driving system, the system comprising an intelligent driving controller and a steering controller of a vehicle, wherein the intelligent driving controller, in the absence of a fault of the intelligent driving controller, exits an assisted driving function based on a hand torque and a duration of the hand torque applied to a steering wheel by a driver, and sends an assisted driving exit instruction to other associated controllers that act on the assisted driving, so as to cause the other associated controllers to exit the assisted driving function; and the steering controller, in the absence of a fault of the steering controller but in the presence of a fault of the intelligent driving controller, exits the assisted driving function based on the hand torque and the duration of the hand torque applied to the steering wheel by the driver, and sends the assisted driving exit instruction to the intelligent driving controller, and the intelligent driving controller, in response to receiving the assisted driving exit instruction sent by the steering controller, exits the assisted driving function and sends the assisted driving exit instruction to the other associated controllers that act on the assisted driving.

[0022] The safety takeover system of the assisted driving system provided by the present application has the advantages that the intelligent driving controller and the steering controller are independent devices, and both can exit the assisted driving function in the absence of a fault of the intelligent driving controller or the steering controller based on the hand torque and the duration of the hand torque applied to the steering wheel by the driver, wherein the intelligent driving controller, in the absence of a fault of the intelligent driving controller, can send the assisted driving exit instruction to the other associated controllers that act on the assisted driving when it is determined that the conditions for exiting the assisted driving function are met, and the steering controller, in the absence of a fault of the steering controller but in the presence of a fault of the intelligent driving controller, sends the assisted driving exit instruction to the intelligent driving controller when it is determined that the conditions for exiting the assisted driving function are met, so as to realize complete takeover by the driver, and no matter whether the intelligent driving controller or the steering controller has a fault, the other device that does not have a fault can still exit the assisted driving function, thereby achieving a double protection effect and ensuring that the driver can successfully take over the vehicle in any case.

[0023] In a fourth aspect, the present application provides a vehicle comprising the safety takeover system of the assisted driving system according to the third aspect or any one of the corresponding embodiments thereof.

[0024] The safety takeover system of the auxiliary driving system provided by the application, the intelligent driving controller and the steering controller are independent devices, and can exit the auxiliary driving function based on the hand torque and duration of the driver acting on the steering wheel in the absence of faults, wherein, the intelligent driving controller can send an auxiliary driving exit instruction to other associated controllers acting on the auxiliary driving in the absence of faults, regardless of whether the steering controller fails or not, when it is determined that the conditions for exiting the auxiliary driving function are met, and the steering controller sends the auxiliary driving exit instruction to the intelligent driving controller when it is determined that the conditions for exiting the auxiliary driving function are met in the case that the intelligent driving controller fails but the steering controller does not fail, so as to realize the complete takeover of the driver, and the other device without faults can still exit the auxiliary driving function when the intelligent driving controller or the steering controller fails, thereby achieving the effect of double protection and ensuring that the driver can successfully take over the vehicle in any case. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the specific embodiments of the application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0026] Figure 1 is a general measure flowchart example diagram of the driver taking over the vehicle according to the embodiment of the application;

[0027] Figure 2 is a structural block diagram of the safety takeover system of the auxiliary driving system according to the embodiment of the application

[0028] Figure 3 is the whole vehicle level safety target information of the IACC cruise function according to the embodiment of the application;

[0029] Figure 4 is the functional safety requirement information of the IACC cruise function according to the embodiment of the application;

[0030] Figure 5 is a safety architecture example diagram of the auxiliary driving system according to the embodiment of the application;

[0031] Figure 6 is a training flowchart example diagram of the threshold prediction model according to the embodiment of the application;

[0032] Figure 7 is an interaction schematic diagram between each port in the safety takeover system of the auxiliary driving system according to the embodiment of the application;

[0033] Figure 8 This is an example diagram of the interaction between various ports in the safety takeover system of the driver assistance system according to an embodiment of the present invention;

[0034] Figure 9 This is a structural block diagram of a vehicle according to an embodiment of the present invention;

[0035] Figure 10 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] During IACC cruise control, under normal vehicle conditions, the driver can typically deactivate the IACC function by turning the steering wheel, pressing the brake pedal, pressing the accelerator pedal, or pressing the exit button. For example... Figure 1 As shown, the method of the driver taking over the vehicle by turning the steering wheel generally involves setting a hand torque threshold (usually around 3 Nm) through the EPS. When the driver's hand torque on the steering wheel exceeds this threshold and the duration exceeds a certain value (usually around 200 ms), the EPS actively disengages the IACC function (i.e., does not execute IACC lateral and longitudinal control commands) and simultaneously sends a driver takeover command to the ADC. Upon receiving the command, the ADC disengages the IACC function and simultaneously sends IACC exit commands to other associated controllers (such as the braking system ESP and the powertrain VCU). All controllers only execute the driver's control commands. This method easily leads to the risk of the ADAS and the driver competing for the steering wheel, and it does not consider the issue of how the driver can safely take over in the event of a malfunction or failure of the steering system (EPS) or intelligent driving controller (ADC) from a functional safety perspective.

[0038] This design takes the Adaptive Cruise Control (IACC) function as an example, and combines the functional characteristics and usage scenarios of IACC. Through Hazard Analysis and Risk Assessment (HARA), the vehicle-level safety objectives of IACC are obtained, aiming to solve the following three problems:

[0039] (1) During IACC cruise, EPS or ADC malfunction or failure can prevent the driver from taking over effectively.

[0040] (2) The failure of EPS and ADC safety mechanisms leads to the driver's inability to take over effectively.

[0041] (3) Under complex working conditions, the driver takeover problem may occur due to unreasonable settings of the torque and duration of the driver takeover.

[0042] This invention provides a safety takeover system for an assisted driving system, such as... Figure 2 As shown, the system includes a vehicle intelligent driving controller 1 and a steering controller 2. When the intelligent driving controller is not faulty, it disengages the driver assistance function based on the driver's hand torque and duration of applying force to the steering wheel, and sends a driver assistance disengagement command to other associated controllers, causing them to disengage their driver assistance functions. When the steering controller detects a fault in the intelligent driving controller, but is not faulty itself, it disengages the driver assistance function based on the driver's hand torque and duration of applying force to the steering wheel, and sends a driver assistance disengagement command to the intelligent driving controller. Upon receiving the driver assistance disengagement command from the steering controller, the intelligent driving controller disengages the driver assistance function and sends driver assistance disengagement commands to other associated controllers.

[0043] The intelligent driving controller and steering controller in this embodiment of the invention are independently operating devices in the vehicle. During vehicle driving, if the user wants to take over the vehicle safely, they can exit the IACC function by turning the steering wheel. Alternatively, if the intelligent driving controller, steering controller, or other related controllers malfunction, the user will be prompted to take over the vehicle safely through the corresponding instrument display, horn, etc., and then the user can exit the IACC function by turning the steering wheel. This is just an example.

[0044] When the vehicle's intelligent driving controller is not faulty, regardless of whether the steering controller is faulty, the first layer of protection mechanism is set. The intelligent driving controller detects the driver's hand torque and duration on the steering wheel through grip force sensor data. If the conditions for disabling the IACC function are met, the intelligent driving controller disabling the driver assistance function will be determined, and a driver assistance disabling command can be sent to other associated controllers that are involved in driver assistance. These other associated controllers may include the steering controller, power system, and braking system, so that they will disabling the driver assistance function upon receiving the driver assistance disabling command. The conditions for disabling the IACC function are not limited. They can be determined that the driver's hand torque on the steering wheel is greater than a preset hand torque threshold, and / or that the duration of the driver's hand torque on the steering wheel is greater than a preset duration threshold.

[0045] The steering controller of the vehicle sets a second protection mechanism in the case that the intelligent driving controller is detected to have a fault but the steering controller itself does not have a fault, detects the steering torque and duration of time that the driver acts on the steering wheel, and exits the assisted driving function if the conditions for exiting the IACC function are met, and sends an assisted driving exit instruction to the intelligent driving controller.

[0046] The safety takeover system of the assisted driving system provided by the application, the intelligent driving controller and the steering controller are independent devices, and can exit the assisted driving function based on the steering torque and duration of time that the driver acts on the steering wheel in the case that the intelligent driving controller and the steering controller themselves do not have faults, wherein the intelligent driving controller can send an assisted driving exit instruction to other associated controllers that act on the assisted driving in the case that the intelligent driving controller itself does not have a fault and the conditions for exiting the assisted driving function are met, and the steering controller sends the assisted driving exit instruction to the intelligent driving controller in the case that the intelligent driving controller is detected to have a fault but the steering controller itself does not have a fault and the conditions for exiting the assisted driving function are met, so as to realize the complete takeover of the driver, and the other device that does not have a fault can still exit the assisted driving function in the case that the intelligent driving controller or the steering controller has a fault, thereby achieving the effect of double protection and ensuring that the driver can successfully take over the vehicle in any case.

[0047] Specifically, the intelligent driving controller and the steering controller are both integrated with corresponding safety mechanisms based on the assisted driving function, and the safety mechanism is used to determine whether the device itself has a fault, wherein the safety mechanism integrated by the steering controller at least includes that the deceleration and deceleration change rate of the steering controller during braking are not greater than a preset first safety threshold, and the steering angle and steering rate of the steering controller during steering are not greater than a preset second safety threshold.

[0048] The embodiment of the application is aimed at the IACC function, performs the concept layer analysis of the whole vehicle level based on the process of the road vehicle functional safety standard GB / T34590, obtains the safety target and safety requirement, and then can combine the possible faults or failures of the assisted driving system to design the corresponding safety mechanism, so as to solve the problems of EPS fault, ADC fault or failure of the EPS and ADC safety mechanism in the IACC cruise process, so that the system enters the corresponding safety state, defines the IACC function, confirms the IACC function requirement, non-functional requirement, interface definition and function boundary, determines the failure mode and whole vehicle level hazard event based on the working scene of the IACC, and then can perform the corresponding risk assessment on each hazard event, such as Figure 3The vehicle safety integrity (ASIL) level of the whole vehicle hazard event is determined, the safety target is derived, and the safety target can be decomposed to obtain functional safety requirements as shown in Figure 4 The functional safety requirements are systematically analyzed, and safety mechanisms related to driver takeover are designed, such as the EPS should send an "unavailable state" to the intelligent driving controller and report a fault when detecting an internal abnormality of the steering system, and the ADC should have a fault diagnosis mechanism for lateral control requests and longitudinal control requests, and when detecting that the requested lateral control or longitudinal control exceeds the safety limit, it should output according to the maximum limit and can prompt the driver to take over the safety of the vehicle, which is only an example to meet the functional safety requirements, so that the system enters the corresponding safety state, thereby reducing the risk of the driver being unable to take over the vehicle to an acceptable range.

[0049] Specifically, for the functional safety problem that the driver cannot take over due to EPS and ADC failure, a system safety architecture as shown in Figure 5 is designed to enable the system to enter the corresponding safety state. The system safety architecture is designed with the following mechanisms: Safety mechanism 1: During IACC cruise, the deceleration and deceleration rate of ESP braking should not exceed the safety threshold. This mechanism can solve the problem that ADC failure or failure causes IACC to unexpectedly issue a brake request or IACC to issue an excessive brake request. This safety mechanism can meet FSR-ESP-1, FSR-ESP-2, FSR-ADC-2, FSR-ADC-3, and FSR-ADC-4 as shown in Figure 4 ; Safety mechanism 2: During IACC cruise, the steering angle and steering rate of EPS steering should not exceed the safety threshold. This mechanism can solve the problem that ADC failure causes IACC to unexpectedly issue a steering request or IACC to issue an excessive steering request. This safety mechanism can meet FSR-EPS-P3, FSR-ADC-3, and FSR-ADC-4, which is only an example.

[0050] Specifically, as shown in Figure 5As shown, the steering controller in the system at least includes a first steering controller and a second steering controller, wherein the first steering controller is the steering controller with the highest priority, and is used to exit the assisted driving function based on the hand torque and duration of time of the driver acting on the steering wheel, and send an assisted driving exit instruction to the intelligent driving controller to make other associated controllers exit the assisted driving function, when the first steering controller does not have a fault; the second steering controller is used to monitor whether the first steering controller has a fault, and if it is monitored that the first steering controller has a fault, the second steering controller updates itself to be the steering controller with the highest priority, and is used to exit the assisted driving function based on the hand torque and duration of time of the driver acting on the steering wheel, and send an assisted driving exit instruction to the intelligent driving controller to make other associated controllers exit the assisted driving function, when the second steering controller does not have a fault; and the second steering controller is also used to send fault information to the intelligent driving controller when it is monitored that the first steering controller has a fault and the second steering controller has a fault.

[0051] The safety mechanism 3 designed in the embodiment of the application includes: the ESP system is redundantly designed, for example, at least two steering controllers can be designed, one is a main steering controller with the highest priority, and the other is an auxiliary steering controller; the main steering controller can exit the assisted driving function based on the hand torque and duration of time of the driver acting on the steering wheel, and send an assisted driving exit instruction to the intelligent driving controller to make other associated controllers exit the assisted driving function, when the main steering controller does not have a fault; and the auxiliary steering controller can detect whether the main steering controller has a fault in real time.

[0052] If the main steering controller has a fault, the EPS steering system still has a certain steering capability, the auxiliary steering controller replaces the step of exiting the assisted driving function based on the hand torque and duration of time of the driver acting on the steering wheel, and sending an assisted driving exit instruction to the intelligent driving controller to make other associated controllers exit the assisted driving function, and if the auxiliary steering controller monitors that the main steering controller has a fault and the auxiliary steering controller has a fault, the auxiliary steering controller can send fault information to the intelligent driving controller, so as to solve the problems that the EPS fault causes the EPS to unexpectedly execute an IACC control request or the driver cannot successfully take over the vehicle and control the vehicle to enter a safe state, and the like, and the safety mechanism can meet the requirements of the FSR-EPS-P4, the FSR-ADC-1, the FSR-ADC-3 and the FSR-ADC-3 as examples. Figure 4 The FSR-EPS-P4, the FSR-ADC-1, the FSR-ADC-3 and the FSR-ADC-3 are only examples.

[0053] Specifically, the hand torque and duration of time of the driver acting on the steering wheel are the result of the driver operating the steering wheel to perform a safe takeover operation, and the driver operating the steering wheel to perform a safe takeover operation is the operation performed by the driver after the intelligent driving controller prompts the driver to perform a vehicle safe takeover in response to receiving fault information sent by the steering controller or detecting that the intelligent driving controller has a fault.

[0054] The steering controller of the embodiment of the present application can detect whether it has a fault by using the integrated safety mechanism designed based on the functional safety requirement, and if it is detected that the steering controller has a fault, the fault information can be sent to the intelligent driving controller, and the intelligent driving controller can prompt the driver to take over the vehicle safely after receiving the fault information sent by the steering controller, so that the driver operates the steering wheel to perform the safe takeover operation, and the intelligent driving controller detects the hand torque and duration of the driver acting on the steering wheel under the condition that the intelligent driving controller itself has no fault, and if the requirement for exiting the IACC function is met, the assisted driving function can be exited, and an assisted driving exit instruction is sent to other associated controllers acting on the assisted driving.

[0055] The intelligent driving controller of the embodiment of the present application can also detect whether it has a fault by using the integrated safety mechanism designed based on the functional safety requirement, and if it is detected that the intelligent driving controller has a fault, the driver can be prompted to take over the vehicle safely, so that the driver operates the steering wheel to perform the safe takeover operation, and the fault information of the intelligent driving controller can be sent to the steering controller, and the steering controller triggers the second layer protection mechanism under the condition that the intelligent driving controller has a fault but the steering controller itself has no fault, that is, the steering controller exits the assisted driving function based on the hand torque and duration of the driver acting on the steering wheel, and sends an assisted driving exit instruction to the intelligent driving controller, so that the intelligent driving controller sends an assisted driving exit instruction to other associated controllers.

[0056] If the steering controller has no fault and the intelligent driving controller also has no fault, the first layer protection mechanism is used, that is, the intelligent driving controller detects the hand torque and duration of the driver acting on the steering wheel, and if the requirement for exiting the IACC function is met, the assisted driving function can be exited, and an assisted driving exit instruction is sent to other associated controllers acting on the assisted driving.

[0057] When the steering controller has a fault, the steering controller can send the fault information to the intelligent driving controller, and the intelligent driving controller can prompt the driver to take over the vehicle safely based on the fault information sent by the intelligent driving controller or the detection that the intelligent driving controller has a fault, so that the driver is timely notified to take over the vehicle when the vehicle has a fault.

[0058] The steering controller and the intelligent driving controller designed in the present application can detect whether they have a fault, and after the fault is detected, the driver is prompted to take over the vehicle safely, and the party without a fault exits the assisted driving function based on the hand torque and duration of the driver acting on the steering wheel, so that the vehicle enters the corresponding safe state.

[0059] Specifically, the intelligent driving controller judges whether the hand torque of the driver acting on the steering wheel is greater than a first hand torque threshold value; if the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold value, a first duration that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold value is recorded; when the first duration is greater than a first duration threshold value, the assisted driving function is exited, and an assisted driving exit instruction is sent to other associated controllers acting on the assisted driving.

[0060] Specifically, the steering controller judges whether the hand torque of the driver acting on the steering wheel is greater than a second hand torque threshold value; if the hand torque of the driver acting on the steering wheel is greater than the second hand torque threshold value, a second duration that the hand torque of the driver acting on the steering wheel is greater than the second hand torque threshold value is recorded; when the second duration is greater than a second duration threshold value, an assisted driving exit instruction is sent to the intelligent driving controller.

[0061] The second hand torque threshold value is greater than the first hand torque threshold value, and the first hand torque threshold value is a threshold value used by the intelligent driving controller to judge whether to exit the assisted driving function based on the hand torque of the driver acting on the steering wheel.

[0062] The embodiment of the application sets two layers of safety mechanism logic, the first layer of safety mechanism logic is that the hand torque is greater than a first hand torque threshold value N1, and the duration is greater than a first duration threshold value, the assisted driving function is exited; the second layer of safety mechanism logic is that the hand torque is greater than a second hand torque threshold value N2, and the duration is greater than a second duration threshold value, the assisted driving function is exited, generally, the first layer of safety mechanism logic is the safety mechanism logic that is first implemented under normal circumstances, the hand torque threshold value set by the first layer of safety mechanism logic can be a smaller threshold value, which facilitates the user to safely take over, when the first layer of safety mechanism logic fails, the second layer of safety mechanism logic is implemented, corresponding to the actual driving process, the driver takes over the vehicle and finds that it cannot be taken over, under normal circumstances, the driver will increase the hand torque value, therefore, the second hand torque threshold value in the second layer of safety mechanism logic is greater than the first hand torque threshold value, N1 is generally about 3 Nm, and N2 can be about 10 Nm, which is only an example.

[0063] Wherein, the smart driving controller of the embodiment of the present application integrates a first layer safety mechanism logic, when the smart driving controller itself does not have a fault, no matter whether the steering controller has a fault, the first layer protection mechanism is triggered, the smart driving controller detects the hand torque and the duration of the hand torque of the driver acting on the steering wheel, judges whether the hand torque of the driver acting on the steering wheel is greater than a first hand torque threshold, recorded as N1, after determining that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold, records the first duration of the hand torque of the driver acting on the steering wheel being greater than the first hand torque threshold, and when the first duration is greater than a first time threshold, recorded as T1, the assisted driving function is exited, and an assisted driving exit instruction is sent to other associated controllers for assisted driving, the reason for the failure of the smart driving controller may be a hardware or software failure of itself, or a failure of the first layer safety mechanism logic integrated by itself, when the first layer safety mechanism logic fails, the present application integrates a second layer safety mechanism logic in the steering controller, realizes the safe takeover of the vehicle, and avoids the problem that the driver cannot take over the vehicle through the steering wheel after the smart driving controller fails.

[0064] The steering controller of the embodiment of the present application integrates a second layer safety mechanism logic, in the case that the smart driving controller fails but the steering controller itself does not have a fault, the second layer protection mechanism is triggered, the steering controller detects the hand torque and the duration of the hand torque of the driver acting on the steering wheel, judges whether the hand torque of the driver acting on the steering wheel is greater than a second hand torque threshold, recorded as N2, and can record the second duration of the hand torque of the driver acting on the steering wheel being greater than the second hand torque threshold, recorded as T2, after determining that the hand torque of the driver acting on the steering wheel is greater than the second hand torque threshold, when the second duration is greater than a second time threshold, an assisted driving exit instruction is sent to the smart driving controller, so that the smart driving controller exits the assisted driving function after receiving the assisted driving exit instruction, and sends an assisted driving exit instruction to the associated controller.

[0065] The present application takes the smart driving controller as the first layer protection mechanism and the steering controller as the second layer protection mechanism, which can solve the problem that the driver cannot take over the vehicle through the steering wheel after the safety mechanism of the smart driving controller or the steering controller fails, has a double protection effect, and ensures that the driver can successfully take over the vehicle in any case. The safety mechanism can meet the requirements of Figure 4 FSR-EPS-P1, FSR-EPS-P2 shown in the figure.

[0066] Further, the steering controller records a third duration that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold when the steering controller determines that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, and sends an assisted driving exit instruction to the intelligent driving controller when the third duration is greater than a third duration threshold.

[0067] The embodiment of the present application considers that the driver wants to take over the vehicle by turning the steering wheel, but the first layer safety mechanism does not work due to the failure of the safety mechanism of the intelligent driving controller, and the hand force value of the driver does not reach the safety threshold of the second layer safety mechanism, which will cause the driver and the assisted driving function to compete for the steering wheel, and the driver will be driven by the steering wheel if the driver does not increase the torque of the steering wheel in time, which will cause safety problems. Therefore, a third layer protection mechanism is set, that is, a third layer safety mechanism logic is designed, wherein the third layer safety mechanism logic is used to exit the assisted driving function when the hand torque is greater than the first hand torque threshold and less than the second hand torque threshold, and the third duration is greater than the third duration threshold. The steering controller integrates the third layer safety mechanism logic, so that the second layer safety logic and the third layer safety logic integrated by the steering controller can be implemented at the same time when the steering controller detects that the intelligent driving controller fails but the steering controller itself does not fail. When the steering controller determines that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, the steering controller records a third duration that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, and the steering controller can send an assisted driving exit instruction to the intelligent driving controller when the third duration is greater than the third duration threshold. The intelligent driving controller exits the assisted driving function after receiving the assisted driving exit instruction, and can send an assisted driving exit instruction to the associated controller.

[0068] The steering controller sends an assisted driving exit instruction to the intelligent driving controller when the steering controller determines that the hand torque of the driver acting on the steering wheel is greater than the first hand torque threshold but less than the second hand torque threshold, and the third duration is greater than the third duration threshold, which avoids the risk that the driver cannot immediately take over the vehicle when the actual hand force value of the driver does not reach the second hand torque threshold.

[0069] Specifically, the intelligent driving controller integrates a threshold prediction model corresponding to different hand torque thresholds and duration thresholds in different driving scenes, obtains a driving scene of the current vehicle, and inputs the driving scene of the current vehicle into the threshold prediction model to obtain a first hand torque threshold and a first duration threshold corresponding to the driving scene of the current vehicle.

[0070] Specifically, the steering controller integrates a threshold prediction model corresponding to different driving scenarios and different hand force thresholds and time length thresholds, obtains the driving scenario of the current vehicle, inputs the driving scenario of the current vehicle into the threshold prediction model, and obtains the second hand force threshold, the second time length threshold, the third hand force threshold and the third time length threshold corresponding to the driving scenario of the current vehicle.

[0071] The intelligent driving controller and the steering controller of the embodiment of the application both integrate a pre-trained threshold prediction model corresponding to different driving scenarios and different hand force thresholds and time length thresholds, wherein the threshold detection model can be based on the vehicle safety takeover situation of historical vehicles in different scenarios to determine the first hand force threshold and the first time length threshold, the second hand force threshold and the second time length threshold, and the third hand force threshold and the third time length threshold, and input them into a neural network model for training to obtain the threshold prediction model. For example only, the intelligent driving controller can obtain the driving scenario of the current vehicle when it is not faulty, wherein the driving scenario at least includes the vehicle state, the road condition and the driver behavior characteristics. For example only, the driving scenario of the current vehicle is input into the threshold prediction model to obtain the first hand force threshold and the first time length threshold corresponding to the driving scenario of the current vehicle, and whether the hand torque and the duration of the current driver acting on the steering wheel meet the condition for the auxiliary driving function to exit is judged based on the first hand force threshold and the first time length threshold.

[0072] When the steering controller is not faulty and detects the hand torque and the duration of the driver acting on the steering wheel, the driving scenario of the current vehicle can be obtained in advance, the driving scenario of the current vehicle is input into the pre-integrated threshold prediction model to obtain the second hand force threshold, the second time length threshold, the third hand force threshold and the third time length threshold corresponding to the driving scenario of the current vehicle, and whether the hand torque and the duration of the current driver acting on the steering wheel meet the condition for the auxiliary driving function to exit is judged based on the second hand force threshold, the second time length threshold, the third hand force threshold and the third time length threshold, for example only.

[0073] The application determines the corresponding hand force threshold and time length threshold based on the driving scenario of the current vehicle, and judges whether the hand torque and the duration of the current driver acting on the steering wheel meet the condition for the auxiliary driving function to exit based on the determined hand force threshold and time length threshold, which can improve the smoothness and safety of the driver takeover.

[0074] Specifically, the threshold prediction model corresponding to different driving scenarios and different hand force thresholds and time length thresholds is obtained through the following steps: based on different data acquisition methods, corresponding safe takeover data under all driving scenarios is respectively acquired, wherein the safe takeover data under any driving scenario includes safe takeover data respectively acquired under different data acquisition methods; according to a preset driving scenario set respectively corresponding to different hand force thresholds and time length thresholds, the acquired safe takeover data under all driving scenarios is respectively configured into the corresponding driving scenario set, and the average hand force torque and the average duration time respectively corresponding to different data acquisition methods under different hand force thresholds and time length thresholds are calculated, wherein the different hand force thresholds and time length thresholds include the first hand force threshold and the first time length threshold, the second hand force threshold and the second time length threshold, and the third hand force threshold and the third time length threshold; the preset weight respectively corresponding to different data acquisition methods under the current driving scenario is weighted and summed with the average hand force torque and the average duration time respectively corresponding to different data acquisition methods under different hand force thresholds and time length thresholds, to obtain the different hand force thresholds and time length thresholds corresponding to the current driving scenario; all driving scenarios are traversed, and the step of weighting and summing the weight respectively corresponding to different data acquisition methods under the current driving scenario with the average hand force torque and the average duration time respectively corresponding to different data acquisition methods under different hand force thresholds and time length thresholds is repeatedly executed to obtain different hand force thresholds and time length thresholds corresponding to different driving scenarios; and the different hand force thresholds and time length thresholds corresponding to different driving scenarios are input into a preset neural network model for training to obtain the threshold prediction model.

[0075] The data acquisition method is not limited in the embodiment of the application, and cloud platform data, real vehicle test data and simulation data can be taken as examples, as shown in the following table: Figure 6 As shown in the table, first, the safe takeover data under all driving scenarios is respectively acquired based on different data acquisition methods, for example, the historical takeover data from different vehicles and different drivers is acquired through the cloud platform, including hand force values, takeover time, vehicle state before takeover, road conditions, driver behavior characteristics, etc., the takeover hand force values, takeover time, vehicle state before takeover, road conditions, driver behavior characteristics, etc. are recorded through real vehicle testing based on specific vehicles and different drivers, and the corresponding simulation data is generated by simulating the takeover process under the cloud platform and real vehicle testing scenarios using advanced assisted driving system simulation software, wherein the simulation data includes takeover hand force values, takeover time, vehicle state before takeover, road conditions, driver behavior characteristics, etc.

[0076] The embodiment of the present application can preprocess the obtained safety takeover data, such as removing outliers and missing values, to ensure the accuracy and reliability of the data, and then extract key features from the preprocessed data, such as vehicle speed, acceleration or deceleration, road curvature, road flatness, driver line of sight direction, and driver gender, just as examples. An adaptive weight allocation mechanism can be designed using machine algorithms (such as support vector machines), and weights can be dynamically allocated to cloud platform data, real vehicle test data, and simulation data in different driving scenarios in real time according to data characteristics, to ensure accurate prediction of driver hand torque and duration in different driving scenarios.

[0077] The embodiment of the present application can pre-set different hand force threshold values and duration threshold values corresponding to different driving scene sets, such as assigning the hand force threshold value to the safety takeover data of the second hand force threshold value in the low-speed driving scene, and assigning the hand force threshold value to the safety takeover data of the first hand force threshold value in the turning driving scene, just as examples. All the obtained safety takeover data in different driving scenes can be assigned to the corresponding driving scene set according to the different hand force threshold values and duration threshold values corresponding to the driving scene set, and the average hand torque and duration of different data acquisition methods under different hand force threshold values and duration threshold values can be calculated, such as the first layer hand torque average values of cloud platform data, real vehicle test data, and simulation data being Ncloud1, Ntest1, and Nsin1, and the duration average values being Tcloud1, Ttest1, and Tcloudsin1, where the first layer safety mechanism represents the first hand force threshold value and the first duration threshold value, cloud represents the cloud platform, test represents the real vehicle test, and sin represents the simulation test. The second layer hand torque average values are Ncloud2, Ntest2, and Nsin2, and the duration average values are Tcloud2, Ttest2, and Tsin2, just as examples. The third layer duration average values are Tcloud3, Ttest3, and Tsin3.

[0078] The embodiment of the application adopts a weighted average method to fuse cloud platform data, real vehicle test data and simulation data, and calculates the hand torque and duration required when the driver takes over in the current scene as the input of the prediction model. The preset weights of different data acquisition methods in the current driving scene are weighted and summed with the hand torque average value and duration average value of different data acquisition methods in different layers, to obtain different hand force thresholds and duration thresholds corresponding to the current driving scene, that is, N=Ncloud*omegacloud+Ntest*omega test+Nsin*omega sin, T=Tcloud*taucloud+Ttest*tau test+Tsin*tau sin, to obtain the first hand force threshold, the first duration threshold, the second hand force threshold, the second duration threshold and the third hand force threshold, and the third duration threshold. All driving scenes are traversed to obtain different hand force thresholds and duration thresholds corresponding to different driving scenes.

[0079] The embodiment of the application constructs a threshold prediction model of dynamic driver takeover hand force threshold and duration threshold based on different hand force thresholds and duration thresholds corresponding to different driving scenes. Suitable machine learning or deep learning algorithms such as neural networks and random forests can be selected to optimize and train the prediction model to obtain the threshold prediction model, so as to ensure the prediction accuracy and generalization ability of the threshold prediction model. In the application stage, the intelligent driving controller and the steering controller can input the first hand force threshold, the first duration threshold, the second hand force threshold, the second duration threshold and the third hand force threshold in real time according to the current vehicle state, road conditions and driver behavior characteristics. The actual takeover situation can also be compared with the data output by the threshold prediction model. If the deviation is within the acceptable range, the current weight can be maintained. If the deviation is outside the acceptable range, the weight of the threshold prediction model can be optimized by using online learning algorithms such as gradient descent method and reinforcement learning.

[0080] The application can allocate weights according to the characteristics of different driving scenes, enhance the adaptability of the system to different driving scenes and drivers, and reduce the safety risks caused by improper takeover.

[0081] The embodiment of the application also provides a safety takeover method of an auxiliary driving system. The method is applied to an intelligent driving controller 1 and a steering controller 2 as shown in Figure 2 As shown in Figure 7 The intelligent driving controller 1 is used to execute steps S101 to S102, and the steering controller 2 is used to execute step S201.

[0082] Step S101, in the case that no fault exists in itself, based on the hand torque and the duration of time that the driver acts on the steering wheel, the assisted driving function is exited, and an assisted driving exit instruction is sent to other associated controllers that act on the assisted driving, so as to make the other associated controllers exit the assisted driving function.

[0083] Step S102, in response to receiving the assisted driving exit instruction sent by the steering controller of the vehicle, the assisted driving function is exited, and an assisted driving exit instruction is sent to other associated controllers that act on the assisted driving, so as to make the other associated controllers exit the assisted driving function.

[0084] Step S201, in the case that no fault exists in itself, based on the hand torque and the duration of time that the driver acts on the steering wheel, an assisted driving exit instruction is sent to the intelligent driving controller of the vehicle.

[0085] In specific embodiments, as shown in Figure 8 , when the driver hand torque is greater than N1 and the duration of time is greater than T1, the ADC exits the IACC, and at the same time, an IACC exit instruction is sent to the associated controller, which serves as the first layer protection mechanism; when the driver hand torque is greater than N2 and the duration of time is greater than T2, the EPS exits the IACC, and at the same time, an IACC exit instruction is sent to the ADC, and after the ADC receives the instruction, the IACC is exited, and at the same time, an IACC exit instruction is sent to the associated controller, which serves as the second layer protection mechanism; when the driver hand torque is greater than N1 and less than N2 and the duration of time is greater than T3, the EPS actively exits the IACC, and at the same time, an IACC exit instruction is sent to the ADC, and after the ADC receives the instruction, the IACC is exited, and at the same time, an IACC exit instruction is sent to the associated controller, which serves as the third layer protection mechanism. For details, please refer to the above-mentioned embodiments, which will not be repeated here.

[0086] The embodiment of the present application also provides a vehicle, as shown in Figure 9 , which comprises the safety takeover system of the assisted driving system described above.

[0087] Please refer to Figure 10 , Figure 10 is a structural schematic diagram of a computer device provided by an optional embodiment of the present application, as shown in Figure 10As shown, the computer device includes one or more processors 10, memory 20, and interfaces 50 for external devices such as modems and network interfaces. The one or more processors 10 can be implemented as one or more central processing units (CPUs), one or more microprocessors, microcontrollers, digital signal processors, specialized processors or controller, or one or more processors of any equivalent known in the art. In some embodiments, the one or more processors 10 can be implemented as a combination of one or more of the above physical machines. The memory 20 can include volatile and nonvolatile memory such as read only memory (ROM) 21, random access memory (RAM) 22, and electrically erasable programmable read only memory (EEPROM) 23. The memory 20 can also include a storage 24, such as a magnetic or optical hard disk, a floppy disk, a CD-ROM, a DVD, a Blu-ray disc, a flash memory, or other suitable storage devices. The memory 20 can include a storage medium configured to store data received from external devices, such as the input device 30, or to store data to be transmitted to external devices, such as the output device 40. The memory 20 can also include a storage medium configured to store instructions executable by the one or more processors 10, such as software or firmware. The memory 20 can include a storage medium configured to store data or instructions that are not executable by the one or more processors 10, such as data files. Figure 10 The processor 10 is taken as an example in the embodiment.

[0088] The processor 10 can be a central processing unit, a network processor or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic gate array, a generic array logic or any combination thereof.

[0089] The memory 20 stores instructions executable by the at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiment.

[0090] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system, application programs required by at least one function, and the like. The data storage area can store data created according to the use of the computer device, and the like. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0091] The memory 20 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk or a solid state disk; the memory 20 can also include a combination of the above kinds of memories.

[0092] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 can be connected by a bus or other means,Figure 10 The bus connections are exemplary only. Other types of connections and media can also be used to connect the chosen devices.

[0093] The input device 30 can receive input digital or character information, and generate key signal input with respect to user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0094] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded through a network and stored in a local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special purpose hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0095] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be suggested by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes are intended to fall within the scope of the appended claims.

Claims

1. A safe takeover method for a driver assistance system, characterized in that, A driving controller for vehicles, the driving controller integrating threshold prediction models for different hand force thresholds and duration thresholds corresponding to different driving scenarios, the method comprising: If there is no fault in itself, based on the torque and duration of the driver's hand force on the steering wheel, the driver assistance function is discontinued, and a driver assistance discontinuation command is sent to other associated controllers that are acting as driver assistance, so that the other associated controllers discontinue the driver assistance function. Alternatively, in response to receiving an assisted driving exit command sent by the vehicle's steering controller, the assisted driving function is exited, and assisted driving exit commands are sent to other associated controllers acting on the assisted driving function, so that the other associated controllers exit the assisted driving function. The assisted driving exit command is an assisted driving exit command sent by the steering controller to the intelligent driving controller when it detects a fault in the intelligent driving controller, but does not have a fault itself, based on the hand torque and duration of the driver's action on the steering wheel. The threshold prediction models for different driving scenarios corresponding to different hand force thresholds and duration thresholds are obtained through the following steps: Based on different data acquisition methods, safety takeover data for all driving scenarios are acquired respectively. Among them, safety takeover data for any driving scenario includes safety takeover data acquired under different data acquisition methods. According to the preset sets of driving scenarios corresponding to different hand force thresholds and duration thresholds, the safety takeover data of all driving scenarios are configured into the corresponding sets of driving scenarios. The average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds are calculated. The different hand force thresholds and duration thresholds include a first hand force threshold and a first duration threshold, a second hand force threshold and a second duration threshold, and a third hand force threshold and a third duration threshold. The weights corresponding to different data acquisition methods under the preset current driving scenario are weighted and summed with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds to obtain the different hand force thresholds and duration thresholds corresponding to the current driving scenario. Traverse all driving scenarios and repeatedly execute the step of weighting and summing the weights corresponding to different data acquisition methods in the current driving scenario with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds, to obtain different hand force thresholds and duration thresholds corresponding to different driving scenarios. Different hand force thresholds and duration thresholds corresponding to different driving scenarios are input into a preset neural network model for training to obtain a threshold prediction model.

2. The method according to claim 1, characterized in that, The torque and duration of the driver's hand force applied to the steering wheel are the result of the driver performing a safe takeover operation by operating the steering wheel. The safe takeover operation by the driver is the operation performed by the driver after the intelligent driving controller prompts the driver to perform a safe takeover of the vehicle in response to receiving fault information sent by the steering controller or detecting a fault in itself.

3. The method according to claim 1, characterized in that, The method of disengaging the driver assistance function based on the driver's hand torque and duration applied to the steering wheel, and sending a driver assistance disengagement command to other associated controllers, includes: Determine whether the driver's hand torque applied to the steering wheel exceeds the first hand force threshold; If the driver's hand torque applied to the steering wheel is greater than the first hand force threshold, record the first duration during which the driver's hand torque applied to the steering wheel is greater than the first hand force threshold; When the first duration exceeds the first duration threshold, the driver assistance function is discontinued, and a driver assistance discontinuation command is sent to other associated controllers that are acting on the driver assistance function.

4. The method according to claim 3, characterized in that, The method further includes: Obtain the current driving scenario of the vehicle; The driving scenario of the current vehicle is input into the threshold prediction model to obtain the first hand force threshold and the first duration threshold corresponding to the driving scenario of the current vehicle.

5. A safe takeover method for a driver assistance system, characterized in that, A steering controller applied to a vehicle, the steering controller integrating a threshold prediction model for different hand force thresholds and duration thresholds corresponding to different driving scenarios, the method comprising: If a fault is detected in the intelligent driving controller, but there is no fault in the controller itself, the driver sends an assisted driving exit command to the intelligent driving controller of the vehicle based on the torque and duration of the driver's hand force on the steering wheel. This causes the intelligent driving controller to exit the assisted driving function based on the assisted driving exit command sent by the steering controller, and sends assisted driving exit commands to other associated controllers that are acting on the assisted driving function, so that the other associated controllers exit the assisted driving function. The threshold prediction models for different driving scenarios corresponding to different hand force thresholds and duration thresholds are obtained through the following steps: Based on different data acquisition methods, safety takeover data for all driving scenarios are acquired respectively. Among them, safety takeover data for any driving scenario includes safety takeover data acquired under different data acquisition methods. According to the preset sets of driving scenarios corresponding to different hand force thresholds and duration thresholds, the safety takeover data of all driving scenarios are configured into the corresponding sets of driving scenarios. The average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds are calculated. The different hand force thresholds and duration thresholds include a first hand force threshold and a first duration threshold, a second hand force threshold and a second duration threshold, and a third hand force threshold and a third duration threshold. The weights corresponding to different data acquisition methods under the preset current driving scenario are weighted and summed with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds to obtain the different hand force thresholds and duration thresholds corresponding to the current driving scenario. Traverse all driving scenarios and repeatedly execute the step of weighting and summing the weights corresponding to different data acquisition methods in the current driving scenario with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds, to obtain different hand force thresholds and duration thresholds corresponding to different driving scenarios. Different hand force thresholds and duration thresholds corresponding to different driving scenarios are input into a preset neural network model for training to obtain a threshold prediction model.

6. The method according to claim 5, characterized in that, The method further includes: The system detects whether it has a fault and sends the fault information to the intelligent driving controller when a fault is detected. The intelligent driving controller then prompts the driver to take over the vehicle safely based on the fault information sent by the steering controller. If the intelligent driving controller itself does not have a fault, it disengages the driver assistance function based on the torque and duration of the driver's hand force on the steering wheel and sends a driver assistance disengagement command to other associated controllers that are using driver assistance functions, so that the other associated controllers disengage the driver assistance functions.

7. The method according to claim 5, characterized in that, The step of disengaging the driver assistance function based on the driver's hand torque and duration applied to the steering wheel, and sending a driver assistance disengagement command to the intelligent driving controller, includes: Determine whether the driver's hand torque applied to the steering wheel exceeds the second hand force threshold; If the driver's hand torque applied to the steering wheel is greater than the second hand force threshold, record the second duration during which the driver's hand torque applied to the steering wheel is greater than the second hand force threshold; When the second duration exceeds the second duration threshold, a driver assistance exit command is sent to the intelligent driving controller. The second hand force threshold is greater than the first hand force threshold. The first hand force threshold is the threshold used by the intelligent driving controller to determine whether to exit the driver assistance function based on the hand torque applied by the driver to the steering wheel.

8. The method according to claim 7, characterized in that, The step of disengaging the driver assistance function based on the driver's hand torque and duration applied to the steering wheel, and sending a driver assistance disengagement command to the intelligent driving controller, also includes: If it is determined that the driver's hand torque applied to the steering wheel is greater than the first hand force threshold but less than the second hand force threshold, record the third duration when the driver's hand torque applied to the steering wheel is greater than the first hand force threshold but less than the second hand force threshold; When the third duration exceeds the third duration threshold, a driver assistance exit command is sent to the intelligent driving controller.

9. The method according to claim 8, characterized in that, The method further includes: Obtain the current driving scenario of the vehicle; The driving scenario of the current vehicle is input into the threshold prediction model to obtain the second hand force threshold, the second duration threshold, the third hand force threshold, and the third duration threshold corresponding to the driving scenario of the current vehicle.

10. A safety takeover system for a driver assistance system, characterized in that, The system includes a vehicle intelligent driving controller and a steering controller. The intelligent driving controller integrates threshold prediction models for different hand force and duration thresholds corresponding to different driving scenarios. The steering controller also integrates threshold prediction models for different hand force and duration thresholds corresponding to different driving scenarios. If the intelligent driving controller is not faulty, it will disengage the driver assistance function based on the torque and duration of the driver's hand force on the steering wheel, and send a driver assistance disengagement command to other associated controllers that are acting as driver assistance controllers, so that the other associated controllers will disengage the driver assistance function. When the steering controller detects a malfunction in the intelligent driving controller, but has no malfunction itself, it disengages the driver assistance function based on the driver's hand torque and duration of applying force to the steering wheel, and sends a driver assistance disengagement command to the intelligent driving controller. Upon receiving the driver assistance disengagement command from the steering controller, the intelligent driving controller disengages the driver assistance function and sends driver assistance disengagement commands to other associated controllers that are implementing driver assistance. The threshold prediction models for different driving scenarios corresponding to different hand force thresholds and duration thresholds are obtained through the following steps: Based on different data acquisition methods, safety takeover data for all driving scenarios are acquired respectively. Among them, safety takeover data for any driving scenario includes safety takeover data acquired under different data acquisition methods. According to the preset sets of driving scenarios corresponding to different hand force thresholds and duration thresholds, the safety takeover data of all driving scenarios are configured into the corresponding sets of driving scenarios. The average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds are calculated. The different hand force thresholds and duration thresholds include a first hand force threshold and a first duration threshold, a second hand force threshold and a second duration threshold, and a third hand force threshold and a third duration threshold. The weights corresponding to different data acquisition methods under the preset current driving scenario are weighted and summed with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds to obtain the different hand force thresholds and duration thresholds corresponding to the current driving scenario. Traverse all driving scenarios and repeatedly execute the step of weighting and summing the weights corresponding to different data acquisition methods in the current driving scenario with the average hand torque and average duration corresponding to different data acquisition methods under different hand force thresholds and duration thresholds, to obtain different hand force thresholds and duration thresholds corresponding to different driving scenarios. Different hand force thresholds and duration thresholds corresponding to different driving scenarios are input into a preset neural network model for training to obtain a threshold prediction model.

11. A vehicle, characterized in that, The vehicle includes a safety takeover system for the driver assistance system as described in claim 10.

Images