Automatic driving manual takeover feedback torque recovery method

Abstract

The invention belongs to the field of automatic driving, and particularly relates to an automatic driving manual takeover feedback torque recovery method. According to the method, on the basis of a force feedback technology, required feedback force is calculated according to information such as the control right and the vehicle state in the takeover process of a driver, so that perception of the driver on the vehicle state and road surface information is enhanced, reaction time is shortened, misoperation caused by stress and tension of the driver is prevented, and the driver experience is improved. Therefore, the vehicle control can be quickly and accurately recovered when the system needs driver intervention. The stability and safety of the pipe connecting process are improved.

Description

Technical Field

[0001] The present invention belongs to the field of autonomous driving, and in particular relates to a method for restoring feedback torque of an autonomous driving manual takeover. Background Art

[0002] With the rapid development of autonomous driving technology, manual takeover has become a key link to ensure the driving safety of L3 autonomous driving vehicles. When the autonomous driving system faces a situation beyond its operational design domain (ODD) or the system fails, control must be returned to the driver in a timely manner. To achieve safe and effective takeover, existing research has focused on optimizing the timing and method of takeover requests, striving to reduce the driver's reaction time and improve the driver's perception of the vehicle status.

[0003] Existing takeover prompt methods mainly include visual prompts, auditory prompts, and tactile prompts. Visual prompts use instruments or HUDs to display warning information, auditory prompts use voice alarms, and tactile prompts use steering wheel shaking to remind the driver to take over control of the vehicle.

[0004] For example, a Chinese patent with publication number CN112722075A discloses a method for taking over the steering wheel of an intelligent driving vehicle, and the steps include: real-time monitoring of whether the electric power steering system is in an activated state; if it is activated, judging whether the hand torque applied to the steering wheel meets the takeover conditions according to the change trend and duration of the hand torque, steering wheel angle and steering wheel angular velocity, and if it meets the conditions, exiting the automatic driving function, and at the same time, the electric power steering system follows the steering demand to assist and take over. That is, by identifying the steering wheel takeover of the steering wheel torque value, the change trend of the steering wheel angle or the steering wheel angular velocity, the driver's intention to take over can be accurately and quickly detected, and the system control right can be released in time to improve safety.

[0005] Another example is a human-machine authority switching method based on steering wheel grip disclosed in a Chinese patent with publication number CN108974013A. The driver's grip is detected by a pressure sensor, and the resistance signal emitted by the pressure sensor is converted into a voltage signal. The voltage signal is read and converted into a corresponding pressure value according to the characteristics of the pressure sensor. The current vehicle control right is detected, and whether the signals emitted by the three pressure sensors meet the switching conditions is determined. Whether the driver is out of the driving state is determined, and whether the auxiliary system should take over the steering wheel. The judgment result of the switching strategy is sent to the auxiliary system controller.

[0006] It can be seen from the above-mentioned disclosed technical solutions that the existing human-machine takeover only solves the problem of whether the driver intends to take over the steering wheel, but ignores the impact of the vehicle status on the takeover effect, which is not conducive to improving the safety of human-machine takeover.

[0007] In addition, during the takeover process, when the autonomous driving system issues a takeover reminder, many current studies are dedicated to optimizing the takeover efficiency from the perspective of the driver's takeover operation and the system's reminder method. For example, when the torque applied by the driver to the steering wheel or the brake pedal travel exceeds the preset threshold, the autonomous driving system will quickly transfer complete control of the vehicle to the driver. However, there are some potential problems in this process: the autonomous driving torque is immediately revoked, causing a sudden change in the torque on the steering wheel, resulting in a deviation in the driver's perception of the vehicle's state, which in turn has an adverse effect on the vehicle's controllability and reduces the stability and safety of the takeover process. Summary of the invention

[0008] The purpose of the present invention is to provide a method for restoring feedback torque during manual takeover of automatic driving. The method is based on force feedback technology and calculates the required feedback force according to information such as the driver's control rights and vehicle status during the takeover process, so as to enhance the driver's perception of the vehicle status and road information, reduce reaction time, and prevent misoperation caused by driver stress and tension, thereby ensuring that vehicle control can be restored quickly and accurately when the system requires driver intervention, thereby improving the stability and safety of the takeover process.

[0009] To achieve the above object, the present invention adopts the following technical solution:

[0010] A method for restoring feedback torque of an automatic driving manual takeover comprises the following steps:

[0011] Step 1: The autonomous driving system issues a takeover request;

[0012] Step 2: judging whether the driver intends to take over based on the signal changes of the steering wheel angle sensor and the torque sensor captured by the automatic driving system, and if so, proceeding to step 3; if not, maintaining automatic driving;

[0013] Step 3: Acquire vehicle status information, calculate the feedback torque transmitted to the driver during manual takeover according to the vehicle status information, and complete the driver's control of the vehicle based on the feedback torque.

[0014] Furthermore, the vehicle state information includes vehicle speed, acceleration and front wheel angle, and the feedback torque delivered to the driver during manual takeover includes the vehicle total self-aligning torque and interaction torque, which is calculated as follows:

[0015] T=T r +T c

[0016] T r =T z-y +T z-z +T f

[0017] T c =k1*(1-α)

[0018] Where T is the feedback torque transmitted to the driver during manual takeover; T r is the total vehicle aligning torque, T c is the interaction torque, when T c =0 means that the control right of the automatic driving system is 0, indicating that the interaction torque disappears completely. At this time, it is considered that the driver has full control over the vehicle; T z-y is the lateral moment, T z-z is the vertical moment, T f is the inherent torque of the steering automatic driving system; k1 is the interaction torque coefficient. Modifying the size of this parameter adjusts the size of the interaction torque transmitted to the driver. It should be noted that the interaction torque adjustment should ensure that the driver can feel the difference between the vehicle and manual driving without affecting the driver's operating sensitivity; (1-α) is the control right of the automatic driving system, which changes with time. When the driver takes over completely, the interaction torque disappears.

[0019] Furthermore, the calculation method of the aligning torque is:

[0020] The formula for calculating the self-aligning moment related to the vertical force is:

[0021] T z-z =-(F flz +F frz )dsinθ kin sinδ f +(F flz -F flz )dsinθ cas cosδ f

[0022] Among them, F flz is the vertical force of the left front wheel, F frz is the vertical force of the right front wheel, θ cas is the caster angle; δ f is the front wheel turning angle, d is the lateral offset of the kingpin, θ kin is the kingpin inclination angle;

[0023] The formula for calculating the lateral force-related self-aligning moment is:

[0024] T z-y =(F fly +F fry )(t n +t c )

[0025] F fly is the lateral force of the left front wheel, Ffry is the lateral force of the right front wheel; t n is the pneumatic tire drag moment; t c is the mechanical drag torque. Furthermore, the calculation method of the interaction torque is:

[0026] Calculate the driver's control weight:

[0027] Define the control weight as: α k+1 =max(0,min(1,α k +αα k )), where α k is the control weight at time k, α k+1 is the control weight at time k+1; Δα k The weight change at time k, Δα k Related to the urgency of takeover and vehicle status, Δα k The calculation formula is:

[0028] αα k =γ·E k ·S k

[0029]

[0030] S k =v k ·a k

[0031] Where γ is the proportionality constant, E k is the urgency of taking over at time k, S k is the vehicle state at time k; d is the obstacle distance, d0 is the threshold distance, β>0, adjustment curve coefficient, v k is the vehicle speed at time k, α k is the acceleration at time k;

[0032] The calculation formula of the interaction torque obtained according to the driver's control weight during the takeover process is T c =k1*(1-α).

[0033] Furthermore, the condition for judging whether the driver has the intention to take over in step 2 is:

[0034] The steering wheel grip is detected for 6 consecutive cycles and reaches the set threshold, or the steering wheel angle changes are detected for 6 consecutive cycles.

[0035] The beneficial effects of the present invention are as follows: the present invention enhances the driver's perception of the vehicle state, reduces reaction time, and prevents misoperation caused by driver stress and tension through an interactive torque based on the driver's control weight in the feedback torque during manual takeover, thereby ensuring that vehicle control can be restored quickly and accurately when the system requires driver intervention, and improving the stability and safety of the takeover process. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The present invention is a flow chart of a method for restoring feedback torque during manual takeover of autonomous driving in an embodiment. DETAILED DESCRIPTION

[0037] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

[0038] like Figure 1 As shown, a flowchart of a method for restoring feedback torque during manual takeover of automatic driving provided in this embodiment includes the following steps:

[0039] Step 1: The autonomous driving system issues a takeover request.

[0040] Step 2: Determine whether the driver intends to take over based on the signal changes of the steering wheel angle sensor and torque sensor captured by the automatic driving system. If yes, proceed to step 3; if no, maintain automatic driving. The conditions for determining whether the driver intends to take over in this embodiment are: monitoring the steering wheel grip force for 6 consecutive cycles and reaching the set threshold, or monitoring the steering wheel angle change for 6 consecutive cycles.

[0041] Step 3: Acquire vehicle status information, calculate the feedback torque transmitted to the driver during manual takeover according to the vehicle status information, and complete the driver's control of the vehicle based on the feedback torque.

[0042] During manual driving, the road has a reaction force on the tires, which generates a self-aligning torque. This self-aligning torque will be transmitted to the driver through a mechanical structure (nowadays, the self-aligning torque can also be simulated by a road sense motor, which is commonly known as "road sense" to transmit to the driver), so that the driver can enhance his perception of road conditions. Without this road sense, the driver will be driving like a bumper car, and in the process of taking over the vehicle, it is difficult for the driver to perceive the control of the vehicle. Therefore, in the process of manual takeover, this embodiment adds an interactive torque to reflect the change in driving weight, and feeds back the change to the driver to enhance the driver's perception. When the control of the vehicle is completely transferred to the driver, the added interactive torque disappears, and only the road sense during manual driving is retained. It has the following description:

[0043] The feedback torque delivered to the driver during manual takeover includes the vehicle's total return torque and interaction torque, which is calculated as follows:

[0044] T=T r +T c

[0045] T r =T z-y +T z-z +T f

[0046] T c =k1*(1-α)

[0047] T=T r +T c

[0048] T r =T z-y +T z-z +T f

[0049] T c =k1*(1-α)

[0050] Where T is the feedback torque transmitted to the driver during manual takeover; T r is the total vehicle aligning torque, T c is the interaction torque, when T c =0 means that the control right of the automatic driving system is 0, indicating that the interaction torque disappears completely. At this time, it is considered that the driver has full control over the vehicle; T z-y is the lateral moment, T z-z is the vertical moment, T f is the inherent torque of the steering automatic driving system; k1 is the interaction torque coefficient. Modifying the size of this parameter adjusts the size of the interaction torque transmitted to the driver. It should be noted that the interaction torque adjustment should ensure that the driver can feel the difference between the vehicle and manual driving without affecting the driver's operating sensitivity; (1-α) is the control right of the automatic driving system, which changes with time. When the driver takes over completely, the interaction torque disappears.

[0051] In this embodiment, the calculation method of the aligning torque is:

[0052] Calculate the self-aligning moment associated with the vertical force:

[0053] T z-z =-(F flz +F frz )dsinθ kin sinδ f +(F flz -F flz )dsinθ cas cosδ f

[0054] Among them, F flz is the vertical force of the left front wheel, F frz is the vertical force of the right front wheel, θ cas is the caster angle; δ f is the front wheel turning angle, d is the lateral offset of the kingpin, θ kin is the kingpin inclination angle;

[0055] Calculate the lateral force-related self-aligning moment:

[0056] T z-y =(F fly +F fry )(t n +t c )

[0057] F fly is the lateral force of the left front wheel, F fry is the lateral force of the right front wheel; t n is the pneumatic tire drag moment; t c Mechanical drag torque.

[0058] In this embodiment, the calculation method of the interaction torque is:

[0059] Calculate the driver's control weight:

[0060] Define the control weight as: α k+1 =max(0,min(1,α k +Δα k )), where α k is the control weight at time k, α k+1 is the control weight at time k+1; Δα k The weight change at time k, Δα k Related to the urgency of takeover and vehicle status, Δα k The calculation formula is:

[0061] Δα k =γ·E k ·S k

[0062]

[0063] S k =v k ·a k

[0064] Where γ is the proportionality constant, E k is the urgency of taking over at time k, S k is the vehicle state at time k; d is the obstacle distance, d0 is the threshold distance, β>0, adjustment curve coefficient, v kis the vehicle speed at time k, α k is the acceleration at time k. In this embodiment, E k The range is 0-1 and d <d0,E k The larger the value, the more urgent the takeover is, and the farther away from the obstacle, the less urgent the takeover is.

[0065] The calculation formula of the interaction torque obtained according to the driver's control weight during the takeover process is T c =k1*(1-α).

[0066] In summary, this embodiment starts from the manual takeover process based on the driver's intention, and uses the vehicle's total self-aligning torque and the interaction torque as the feedback torque in the manual takeover process as feedback conditions, gradually transferring the control right from the automatic driving system to the driver. This enables the driver to quickly and accurately restore vehicle control when the system requires driver intervention, effectively improving the stability and safety of the takeover process.

[0067] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for restoring feedback torque during manual takeover of an automatic driving, characterized in that: The following steps are involved: Step 1: The autonomous driving system issues a takeover request; Step 2: judging whether the driver intends to take over based on the signal changes of the steering wheel angle sensor and the torque sensor captured by the automatic driving system, and if so, proceeding to step 3; if not, maintaining automatic driving; Step 3: Acquire vehicle status information, calculate the feedback torque transmitted to the driver during manual takeover according to the vehicle status information, and complete the driver's control of the vehicle based on the feedback torque.

2. The method for restoring feedback torque of automatic driving manual takeover according to claim 1, characterized in that: The vehicle status information includes vehicle speed, acceleration and front wheel angle. The feedback torque delivered to the driver during manual takeover includes the vehicle total self-aligning torque and interaction torque, which are calculated as follows: T=T r +T c T r =T z-y +T z-z +T f T c =k1*(1-a) Where T is the feedback torque transmitted to the driver during manual takeover; T r is the total vehicle aligning torque, T c is the interaction torque, when T c =0 means that the control right of the automatic driving system is 0, indicating that the interaction torque disappears completely. At this time, it is considered that the driver has full control over the vehicle; T z-y is the lateral moment, T z-z is the vertical moment, T f is the inherent torque of the steering automatic driving system; k1 is the interaction torque coefficient. Modifying the size of this parameter adjusts the size of the interaction torque transmitted to the driver. It should be noted that the interaction torque adjustment should ensure that the driver can feel the difference between the vehicle and manual driving without affecting the driver's operating sensitivity; (1-α) is the control right of the automatic driving system, which changes with time. When the driver takes over completely, the interaction torque disappears.

3. The method for restoring feedback torque of automatic driving manual takeover according to claim 2, characterized in that: The calculation method of the aligning torque is: The formula for calculating the self-aligning moment related to the vertical force is: T z-z =-(F flz +F frz )dsinθ kin sinδ f +(F flz -F flz )dsinθ cas cosδ f Among them, F flz is the vertical force of the left front wheel, F frz is the vertical force of the right front wheel, θ cas is the caster angle; δ f is the front wheel turning angle, d is the lateral offset of the kingpin, θ kin is the kingpin inclination angle; The formula for calculating the self-aligning moment related to the lateral force is: T z-y =(F fly +F fry )(t n +t c ) F fly is the lateral force of the left front wheel, F fry is the lateral force of the right front wheel; t n is the pneumatic tire drag moment; t c Mechanical drag torque.

4. The method for restoring feedback torque of automatic driving manual takeover according to claim 3, characterized in that: The calculation method of the interaction torque is: Calculate the driver's control weight: Define the control weight as: α k+1 =max(0,min(1,α k +Δα k )), where α k is the control weight at time k, α k+1 is the control weight at time k+1; Δα k The weight change at time k, Δα k Related to the urgency of takeover and vehicle status, Δα k The calculation formula is: Da k =γ·E k ·S k S k =v k ·a k Where γ is the proportionality constant, E k is the urgency of taking over at time k, S k is the vehicle state at time k; d is the obstacle distance, d0 is the threshold distance, β>0, adjustment curve coefficient, v k is the vehicle speed at time k, α k is the acceleration at time k; The calculation formula of the interaction torque obtained according to the driver's control weight during the takeover process is T c =k1*(1-α).

5. The method for restoring feedback torque of an automatic driving manual takeover according to any one of claims 1 to 4, characterized in that: The conditions for judging whether the driver intends to take over in step 2 are: The steering wheel grip is detected for 6 consecutive cycles and reaches the set threshold, or the steering wheel angle changes are detected for 6 consecutive cycles.

Images