Steering wheel out-of-hand detection method, system and equipment and medium

Abstract

The invention discloses a method, a system and equipment for detecting hand release of a steering wheel and a medium. The method comprises the following steps: considering an inertia effect and a damping effect of a steering wheel, and establishing a kinetic model of an automobile steering system; the driver hand torque is observed through a driver hand torque observer, the driver hand torque serves as a state variable, a state equation of the automobile steering system is established in combination with a kinetic model of the automobile steering system, and the steering angle of the steering wheel is obtained according to the state equation of the automobile steering system and a super-spiral sliding mode control algorithm. Establishing a driver hand torque observer; and comparing the torque threshold value and the time threshold value with the observed hand torque of the driver to obtain a hand release detection result. According to the method, the inertia effect and the damping effect of the steering wheel are considered, the influence of the interference torque on the observation result is reduced, the more accurate driver hand torque can be obtained, and the adopted observer can effectively restrain the chattering phenomenon of a traditional sliding mode observer.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile safe driving, and in particular to a method, system, device and medium for detecting hands-off steering wheel. Background Art

[0002] The Society of Automotive Engineers (SAE) divides the degree of vehicle automation into five stages. Only when the vehicle reaches the highest level of automation will a human driver no longer be needed to operate the vehicle. At present, the autonomous driving technology promoted by various car companies is generally at the L2 stage. The L2 stage belongs to the automation level of assisted driving, and the driver is the leader and responsible person for driving. Under the vehicle's intelligent assisted driving conditions, the driver's hands should not leave the steering wheel. In order to avoid the potential risks caused by the driver's hands leaving the steering wheel due to special reasons or bad habits, it is of great significance to develop an accurate and timely steering wheel hands-off detection function.

[0003] Hands-off steering wheel detection is a key technology to ensure driving safety. It needs to be able to correctly identify the state of the driver's hands operating the steering wheel, so that when certain conditions are met, the driver can be alerted by sound or light, the assisted driving function can be exited, or even forced to pull over. When an emergency occurs, the driver actively controls the steering wheel, and the system needs to be able to promptly identify the driver's control intention and return the control to the driver.

[0004] There are three mainstream steering wheel solutions in the industry: (1) Capacitive sensors, pressure sensors, etc. are integrated into the steering wheel to directly detect whether the driver's hands are on the steering wheel. This solution requires additional sensors, which are difficult to calibrate and are easily interfered with; (2) Rely on computer vision algorithms to detect, locate and track the driver's hands. This solution has higher accuracy, but is also more expensive and difficult, and also faces issues of personal privacy and security. (3) Compare the steering wheel torque with the preset standard value to detect the driver's hands-off state. This solution has high requirements for the torque sensor of the steering system, and is easily affected by driving habits and external environment, with a high misjudgment rate.

[0005] In the patent application JP2017-206071A of Toyota Motor Corporation, a calculation method for obtaining the hand torque applied by the driver to the steering wheel is proposed. This method calculates the inertia torque and damping torque of the steering wheel and steering column through the steering wheel angle sensor in the steering system, obtains the torsion bar torque through the torque sensor, and adds the three torques to obtain the driver's hand torque. This method requires the addition of an additional steering wheel angle sensor, and has high requirements for the real-time and synchronization of the sensor. The calculation process lacks a feedback adjustment mechanism. Using the driver's hand torque obtained by this method for steering wheel hands-off detection will result in low reliability. Summary of the invention

[0006] To solve at least one of the problems existing in the prior art, the present invention provides a method, system, device and medium for detecting steering wheel detachment. By using a driver's hand torque observer to estimate the driver's hand torque to determine the state of the driver's operation of the steering wheel, no additional sensors are required, and the accuracy is higher than that of general detachment detection methods based on torque sensors.

[0007] To achieve the object of the present invention, a method for detecting steering wheel detachment provided by the present invention includes the following steps:

[0008] Considering the inertia effect and damping effect of the steering wheel, establish a dynamic model of the vehicle steering system;

[0009] Obtain the torsion bar torque and the steering wheel angle:

[0010] Obtain the driver's hand torque through a driver's hand torque observer. Among them, taking the driver's hand torque as the state variable, combine the dynamic model of the vehicle steering system to establish the state equation of the vehicle steering system, and establish the driver's hand torque observer according to the state equation of the vehicle steering system and the super-twisting sliding mode control algorithm;

[0011] Compare the torque threshold and the time threshold with the observed driver's hand torque to obtain the detachment detection result.

[0012] Further, when establishing the dynamic model of the vehicle steering system, taking the Column Electric Power Steering System (CEPS) as the research object, establish the dynamic models of the steering column and the torque sensor respectively. When the driver drives the vehicle and turns the steering wheel, a rotation angle is input to the input shaft of the torsion bar, and the torsion bar generates a torque due to the difference in the upper and lower rotation angles. The steering system determines how to output the assist torque according to the torque signal and the vehicle speed signal. The steering dynamic model of the CEPS system specifically includes:

[0013] Steering column:

[0014] Torque sensor: K(θ sw -θ e )=T tb (2)

[0015] Among them, J sw is the moment of inertia of the steering wheel, B sw is the steering wheel damping coefficient, K is the torsion bar stiffness, θ sw is the steering wheel angle (torsion bar input shaft angle), θ e is the pinion angle (torsion bar output shaft angle), T d is the driver's hand torque, T tbis the output value of the torque sensor (torque of the torsion bar).

[0016] Further, in the process of obtaining the torsion bar torque and the steering wheel angle, a combination of a TOS sensor and an angle encoder is used to obtain the torsion bar torque and the steering angle of the assist motor respectively, and then the steering wheel angle is calculated by the transmission ratio between the motor angle and the steering wheel angle; or, the torsion bar torque and the steering wheel angle are obtained simultaneously through a TAS sensor.

[0017] Compared with the general sliding mode observer algorithm, the super-twisting sliding mode observer can significantly suppress the chattering of the observation results. In addition, when designing the driver's hand torque observer, the inertial effect and damping effect of the steering wheel are considered, which can effectively reduce the influence of interference torque and the like on the observation of the driver's hand torque.

[0018] Two calibration quantities, a torque threshold and a time threshold, are introduced in the determination of hands-off detection. The specific process includes: comparing the observed driver's hand torque with the torque threshold in each detection cycle. If the driver's hand torque is greater than the torque threshold, the counter value is increased by a preset value. When the count value of the counter reaches the maximum value, it means that the duration has reached the time threshold for the hold determination, and it is determined that the driver is holding the steering wheel with both hands; if the driver's hand torque is less than the torque threshold, the counter is decreased by a preset value in each counting cycle. When the count value of the counter reaches the minimum value, it means that the duration has reached the time threshold for the hands-off determination, and it is determined that the driver has left the steering wheel with both hands. Otherwise, the determination result is the same as that of the previous cycle.

[0019] The torque threshold can be calibrated in experiments: for the torque threshold, due to reasons such as road surface bumps, zero drift of the torque sensor, and residual friction of the torque sensor, even when the driver leaves the steering wheel with both hands, the output of the torque sensor may deviate from zero. Therefore, it is necessary to calibrate the corresponding torque threshold under different vehicle speeds, loads, and road conditions, so that under different working conditions, by comparing the magnitude relationship between the observed driver's hand torque and the torque threshold, the state of the driver operating the steering wheel with both hands can be distinguished as accurately as possible.

[0020] The time threshold can be calibrated according to regulations or driving experience: for example, under different driving conditions, when the steering wheel switches from the "hands-off" state to the "hold" state or from the "hold" state to the "hands-off" state, the time threshold used for the hands-off detection function to identify can be set to different values.

[0021] The characteristics of the counter are as follows: In each counting cycle, the preset value increased by the counter should be greater than the preset value subtracted, because the hand-off detection requires quick recognition when the driver takes over the steering wheel; when the driver's hands leave the steering wheel, to avoid misjudgment, it is necessary that the state where the driver's hand force is less than the torque threshold lasts for a sufficient long time before "hand-off" is confirmed. The confidence level can be calculated based on the value of the counter at the current moment. The closer the count value is to the maximum value, the higher the confidence level of "holding" is considered, and vice versa, the higher the confidence level of "hand-off" is considered.

[0022] A steering wheel hand-off detection system provided by the present invention includes the following modules:

[0023] A dynamic model establishment module for establishing a dynamic model of the vehicle steering system;

[0024] A sensor signal acquisition module for acquiring the torsion bar torque and the steering wheel angle:

[0025] A driver hand torque observer observation module for observing the driver hand torque through a driver hand torque observer. Among them, taking the driver hand torque as the state variable, combining the dynamic model of the vehicle steering system to establish the state equation of the vehicle steering system, and establishing the driver hand torque observer according to the state equation of the vehicle steering system and the super-twisting sliding mode control algorithm;

[0026] A hand-off detection module for comparing the torque threshold and the time threshold with the observed driver hand torque to obtain the hand-off detection result.

[0027] A computer device provided by the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. It is characterized in that when the processor executes the computer program, the described steering wheel hand-off detection method is implemented.

[0028] A computer-readable storage medium provided by the present invention stores a computer program. It is characterized in that when the computer program is executed by a processor, the described steering wheel hand-off detection method is implemented.

[0029] Compared with the prior art, the present invention can at least achieve the following beneficial effects:

[0030] 1) When establishing the dynamic model, the inertial effect and damping effect of the steering wheel are considered, reducing the influence of the interference torque on the observation result, and a more accurate driver hand torque can be obtained.

[0031] 2) A super-twisting sliding mode observer is established to observe the driver hand torque. This observer has good robustness and can effectively suppress the chattering phenomenon of the traditional sliding mode observer.

[0032] 3) During the determination process of the release detection, the torque threshold and the time threshold can be calibrated based on tests, so as to adapt to different working conditions and reduce the possibility of misjudgment in the release detection. Description of the Drawings

[0033] Figure 1 It is a schematic structural diagram of the CEPS steering system in an embodiment of the present invention.

[0034] Figure 2 It is a schematic diagram of the simplified steering column and torque sensor model in an embodiment of the present invention.

[0035] Figure 3 It is a structural block diagram of the super-twisting sliding mode observer in an embodiment of the present invention. In the figure, represents the operation of taking the absolute value and then taking the square root, x represents the multiplication operation, S represents the differentiation of the incoming data, represents the integration operation.

[0036] Figure 4 It is a flow chart of the release detection determination logic in an embodiment of the present invention.

[0037] Figure 5 It is a comparison chart of the torsion bar torque and the observed driver's hand torque obtained from the vehicle test in an embodiment of the present invention.

[0038] Figure 6 It is a comparison chart of the real steering wheel operation detection state and the release detection result obtained from the vehicle test in an embodiment of the present invention. Detailed Embodiment

[0039] To make the purpose, technical solutions and advantages of the present invention clearer and more definite, the following takes embodiments in conjunction with the drawings to further describe the present invention in detail.

[0040] A method for detecting the release of a steering wheel based on estimating the driver's hand torque by a super-twisting sliding mode observer provided by the present invention includes the following steps:

[0041] Step 1: Considering the inertia effect and damping effect of the steering wheel, establish a dynamic model of the vehicle steering system. Subsequently, based on this mathematical model, a super-twisting sliding mode observer will be designed.

[0042] Taking the column electric power steering system (CEPS) as the research object, establish the dynamic models of the steering column and the torque sensor respectively. As Figure 1As shown in the figure, in the CEPS system, when the driver turns the steering wheel, an angle is input to the input shaft of the torsion bar. Due to the angular difference between the upper and lower parts, the torsion bar generates torque. The torque sensor calculates the torsion bar torque using the Hall effect. Combining with the vehicle speed signal, the ECU can determine the magnitude of the output assist torque and control the motor to output the corresponding torque. The torque output by the motor is amplified by the reducer and applied to the output shaft. The pinion on the output shaft drives the rack to move, realizing the steering function of the vehicle.

[0043] The CEPS system is further simplified into a steering column and torque sensor model, as Figure 2 shown. The following is the establishment of the mathematical equation, that is, the dynamic model of the vehicle steering system:

[0044] Steering column:

[0045] Torque sensor: K(θ sw -θ e ) = T tb (2)

[0046] where J sw is the moment of inertia of the steering wheel, B sw is the damping coefficient of the steering wheel, K is the stiffness of the torsion bar, θ sw is the steering wheel angle (the angle of the input shaft of the torsion bar), θ e is the pinion angle (the angle of the output shaft of the torsion bar), T d is the driver's hand torque, T tb is the output value of the torque sensor (the torsion bar torque), represents the inertial torque caused by the inertial effect of the steering wheel, represents the damping torque caused by the damping effect of the steering wheel. J sw 、B sw 、K can be obtained by referring to the relevant materials of the CEPS product or through experimental measurement.

[0047] When the steering wheel is stationary or rotating at a very low speed, the torque value T tb detected by the torque sensor is equal to the driver's hand torque T d ; when the steering wheel is in an obvious dynamic state, the difference between the two cannot be ignored. Therefore, an observer is designed according to formula (1) and formula (2).

[0048] Step 2: Obtain relevant sensor signals: Obtain the torsion bar torque and the assist motor angle from the torque sensor and the assist motor angle encoder in the CEPS system respectively, and calculate the steering wheel angle based on the transmission ratio between the assist motor and the steering wheel; utilize the built-in sensors in the CEPS system.

[0049] In some embodiments of the present invention, a combination of a TOS sensor and an angle encoder can be employed to separately obtain the torsion bar torque and the power assist motor rotation angle, and then the steering wheel rotation angle can be deduced through the transmission ratio between the motor rotation angle and the steering wheel rotation angle; alternatively, only the TAS sensor can be used to simultaneously obtain the torsion bar torque and the steering wheel rotation angle. That is, a suitable sensor solution can be selected according to the existing equipment conditions or cost to obtain the required signals.

[0050] Step 3: Observe the driver's hand torque through a driver's hand torque observer. Among them, taking the driver's hand torque as the state variable, combining the dynamic model of the vehicle steering system to establish the state equation of the vehicle steering system, and based on the state equation of the vehicle steering system and the super-twisting sliding mode control algorithm, the driver's hand torque observer is established.

[0051] Driver's hand torque T d is an unknown input in the dynamic model of the vehicle steering system. Regarding it as a state variable, based on the state equation of the CEPS steering system and the super-twisting sliding mode control algorithm, a driver's hand torque observer is established. In some embodiments of the present invention, the stability of the observer is verified according to the Lyapunov second method, and the value ranges of the required observer parameters (including c (constant), γ (constant), g (feedback gain coefficient), coefficients k1, k2) are obtained to ensure that the driver's hand torque observer can accurately observe the driver's hand torque.

[0052] The structural block diagram of the driver's hand torque observer is as Figure 3 shown. The driver's hand torque observer is a super-twisting sliding mode observer. The specific design method of the super-twisting sliding mode observer is as follows:

[0053] Regarding the driver's hand torque T d as a state variable, and considering that the sampling period T s of the torque sensor is relatively small. It is generally considered that within the sampling period T s the driver's hand torque T d has no change. Therefore, it is considered that Combining the dynamic model of the vehicle steering system, the state equation of the CEPS steering system can be expressed by the following formula:

[0054]

[0055] where t is time and ω is the angular velocity of the steering wheel rotation.

[0056] The design of the driver's hand torque observer based on the super-twisting sliding mode algorithm is as follows:

[0057]

[0058] Among them, U is the super-twisting sliding mode control law, P is a function of U, representing the compensation amount of the observed value of the steering wheel rotation speed, g is the feedback gain coefficient, is the estimated value of the steering wheel angle, is the estimated value of the angular velocity of the steering wheel rotation, is the estimated value of the driver's hand torque. is the derivative of the angular velocity of the steering wheel rotation, is the estimated value of the derivative of the steering wheel angle.

[0059] Let the estimated error of the steering wheel angle be e1, the estimated error of the steering wheel rotation speed be e2, and the estimated error of the driver's hand torque be e3. Then there are:

[0060]

[0061] Design the sliding surface as:

[0062]

[0063] Among them, c is a constant. In some embodiments of the present invention, c>0 can be taken.

[0064] Since the solution of s = 0 is:

[0065]

[0066] As long as s = 0 is set, it can be ensured that the errors e1 and e2 can decrease exponentially to the vicinity of 0. The larger c is, the faster the error converges.

[0067] In order to ensure that the system can move towards the trend of s = 0 during the sliding mode motion, it is necessary to analyze the derivative of s, and then select an appropriate control law u.

[0068]

[0069] The function P can be designed as:

[0070]

[0071] Among them, γ is a constant.

[0072] Then:

[0073]

[0074] When the system moves to the sliding surface, it satisfies At this time, there is:

[0075]

[0076] For the above equation, since J sw > 0, in order to ensure that the estimation error of the driver's hand torque can quickly tend to 0, we can set:

[0077]

[0078] The reaching law of the super-twisting sliding mode algorithm can generally be expressed as:

[0079]

[0080] where is a discontinuous differential term introduced by the super-twisting sliding mode algorithm. The variable v is an auxiliary control variable and an introduced intermediate variable without actual physical meaning. The role of v is to help the system slide stably on the sliding surface and further suppress the influence of external disturbances and unmodeled dynamics. The introduction of v is equivalent to adding an integral link to the control law, which makes the super-twisting sliding mode control law similar to a proportional-integral (PI) controller to a certain extent and can significantly reduce the high-frequency jitter of the control input.

[0081] where k1 and k2 are design parameters and are greater than zero, and d is a bounded disturbance of the system. According to the above analysis, we can obtain:

[0082]

[0083] where δ represents the upper limit of the absolute value of the bounded disturbance and can take a value greater than |d|.

[0084] Based on the above analysis, select the reaching rate:

[0085]

[0086] Combining Equation (9) and Equation (12), the control law can be determined as follows:

[0087]

[0088] To verify the stability of the observer and find the value range of k1 and k2, select the Lyapunov function as:

[0089]

[0090] Set:

[0091]

[0092] Then:

[0093]

[0094] where:

[0095]

[0096] Substituting the disturbance boundary condition, i.e., formula (13), we can obtain:

[0097]

[0098] Where:

[0099]

[0100] To make It is necessary to satisfy Q>0 to ensure the global asymptotic stability of the system, that is, the gain needs to satisfy at this time:

[0101]

[0102] It can be seen from the above formula that when the coefficients k1 and k2 are selected as relatively large numbers, the stability of the system can be guaranteed.

[0103] Compared with the general sliding mode observer algorithm, the super-twisting sliding mode observer can significantly suppress the chattering of the observation results. In addition, when designing the driver's hand torque observer, the inertial effect and damping effect of the steering wheel are considered, which can effectively reduce the influence of interference torque and the like on the observation of the driver's hand torque.

[0104] The present invention regards the driver's hand torque as an unknown input as a state variable, takes the torsion bar torque and the steering wheel angle as input variables, establishes a super-twisting sliding mode observer, adjusts the observer parameters to ensure the stability of the observer and the observation error can converge to 0, so as to observe the accurate driver's hand torque.

[0105] Step 4: Off-hand detection determination: Compare the torque threshold and the time threshold with the observed driver's hand torque to obtain the off-hand detection result.

[0106] The flowchart of the off-hand detection determination logic is as Figure 4 shown. Two calibration quantities are introduced, namely the torque threshold and the time threshold. If the driver's hand torque observed by the driver's hand torque observer is less than the torque threshold, and the holding time of the state where the driver's hand torque is less than the torque threshold is greater than the time threshold, it is considered that the driver's both hands have left the steering wheel, and the "off-hand" signal is output; if the observed driver's hand torque is greater than the torque threshold, and the holding time of the state where the driver's hand torque is greater than the torque threshold is greater than the time threshold, it is considered that the driver's both hands are holding the steering wheel, and the "holding" signal is output; otherwise, the original off-hand detection result remains unchanged.

[0107] In some embodiments of the present invention, in step 4, the torque threshold can be calibrated in experiments: for the torque threshold, due to reasons such as road surface bumps, zero drift of the torque sensor, and residual friction of the torque sensor, even when the driver's hands leave the steering wheel, the output of the torque sensor may deviate from zero. Therefore, in some embodiments of the present invention, through experimental tests under different vehicle speeds, loads, and road conditions, the corresponding torque thresholds are calibrated to form a torque threshold calibration table. Subsequently, the torque threshold under the corresponding working conditions can be conveniently confirmed by looking up the table, so that under different working conditions, by comparing the magnitude relationship between the observed driver's hand torque and the torque threshold, the state of the driver's hands operating the steering wheel can be distinguished as accurately as possible.

[0108] In other embodiments, after calibrating the torque thresholds under different vehicle speeds, loads, and road conditions through experimental tests, a mathematical relationship can be established between information such as vehicle speed, load, road conditions, and the corresponding torque thresholds. Subsequently, the torque threshold under the corresponding working conditions can be obtained through this mathematical relationship.

[0109] In some embodiments of the present invention, in step 4, the time threshold can be calibrated according to regulations or driving experience. For example, in different driving conditions, when the steering wheel switches from the "off - hand" state to the "held" state or from the "held" state to the "off - hand" state, the time threshold of the off - hand detection function can be set to different values. In addition, under the same working conditions, the time threshold for the steering wheel to switch from the "off - hand" state to the "held" state should be less than the time threshold for switching from the "held" state to the "off - hand" state. The time thresholds at different vehicle speeds can be obtained through experiments to form a time threshold calibration table. Subsequently, the time threshold can be directly determined by looking up the table. In other embodiments, a mathematical relationship between vehicle speed and time threshold can be established based on the calibrated different vehicle speeds and time thresholds. Subsequently, the time threshold can be calculated and determined through this mathematical relationship.

[0110] In this step, in each detection cycle, the magnitudes of the observed driver's hand torque and the torque threshold are compared. If the driver's hand torque is greater than the torque threshold, the counter value is incremented by a preset value. When the count value of the counter reaches the maximum value, it means that the duration has reached the time threshold for the "held" determination, and it is determined that the driver is holding the steering wheel with both hands; if the driver's hand torque is less than the torque threshold, the counter is decremented by a preset value in each detection cycle. When the count value of the counter reaches the minimum value, it means that the duration has reached the time threshold for the "off - hand" determination, and it is determined that the driver's hands have left the steering wheel. Otherwise, the determination result is the same as that of the previous detection cycle.

[0111] The preset value added by the counter each time should be greater than the preset value subtracted. This is because the hands-off detection requires rapid identification when the driver takes over the steering wheel; and when the driver's hands leave the steering wheel, in order to avoid misjudgment, the result will only be finally confirmed when the driver's hand torque is less than the torque threshold for a time threshold. In addition, the confidence level can be calculated based on the value of the counter at the current moment. The closer the count value is to 50, the higher the confidence level of "holding", and vice versa.

[0112] In some embodiments of the present invention, 0.02s is set as a detection cycle. When the steering wheel switches from a "hands-off" state to a "hands-on" state, it is necessary to identify the "hands-on" state after 0.2s (time threshold). When the observed driver's hand torque is greater than the torque threshold, the counter adds a preset value (set to 5) in each detection cycle. After 10 detection cycles, the counter reaches a maximum value (set to 50), that is, when the counter reaches the maximum value after 0.2s, it can be determined that the driver is holding the steering wheel with both hands. Similarly, if the steering wheel switches from a "hands-off" state, it takes 2s (time threshold) to identify the "hands-off" state, when the driver's hand torque is less than the torque threshold, the counter subtracts a preset value (set to 0.5) in each detection cycle. After 100 cycles, when the counter count value reaches the minimum value (set to 0), that is, when the counter reaches the minimum value after 2s, it can be determined that the driver has left the steering wheel with both hands.

[0113] Step 5: Actual vehicle test.

[0114] In order to evaluate the performance of the hands-off steering wheel detection HOD, in some embodiments of the present invention, the actual vehicle test conditions are set as follows: the vehicle is driving in a straight line on a non-absolutely flat road, and the driver repeatedly holds and releases the steering wheel in a cycle of 5 seconds until the experiment is completed. Figure 5 It can be seen that when the driver holds the steering wheel, the observed driver hand torque is close to the torsion bar torque detected by the torque observer; when the driver lets go of the steering wheel, the torsion bar torque fluctuates significantly due to factors such as road interference, but the observed driver hand torque is almost 0. Therefore, the method proposed in the present invention has good robustness.

[0115] like Figure 6 As shown, when the driver switches the state of steering wheel operation from "holding" to "hands off", the hands-off detection will output the result of "hands off" in 2s. This is to avoid misjudgment of "hands off" as much as possible; when the driver switches the state of steering wheel operation from "hands off" to "holding", the hands-off detection will almost immediately output the result of "holding". This is to promptly identify the driver's intention to take over in an emergency, thereby returning the steering wheel control to the driver.

[0116] In some embodiments of the present invention, a steering wheel detachment detection system is provided for implementing the steering wheel detachment detection method provided in the foregoing embodiments. The system includes the following modules:

[0117] A dynamic model establishment module for establishing a dynamic model of an automotive steering system;

[0118] A sensor signal acquisition module for acquiring torsion bar torque and steering wheel angle:

[0119] A driver hand torque observer observation module for observing the driver hand torque through a driver hand torque observer. Among them, taking the driver hand torque as a state variable, combining the dynamic model of the automotive steering system to establish a state equation of the automotive steering system, and establishing the driver hand torque observer according to the state equation of the automotive steering system and the super-twisting sliding mode control algorithm;

[0120] A detachment detection module for comparing a torque threshold and a time threshold with the observed driver hand torque to obtain a detachment detection result.

[0121] Among them, a torque threshold confirmation module and a time threshold confirmation module are further included. The torque threshold confirmation module is used to obtain the torque threshold, and the time threshold confirmation module is used to obtain the time threshold.

[0122] In some embodiments of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steering wheel detachment detection method provided in the foregoing embodiments is implemented.

[0123] In some embodiments of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steering wheel detachment detection method provided in the foregoing embodiments is implemented.

[0124] The embodiments of the present invention take the CEPS system equipped with a torque sensor as the research object, observe the accurate driver hand torque through a super-twisting sliding mode observer, and combine the driver hand torque information with the torque threshold and the time threshold to determine whether the state of the driver operating the steering wheel is "detached" or "held". Compared with the prior art, the embodiments of the present invention have no additional hardware costs; the influence of the inertia effect and damping effect of the steering wheel on the detachment detection is considered during modeling. In addition, the method of observing the driver hand torque by using a super-twisting sliding mode observer has strong robustness and suppresses the chattering phenomenon of a general sliding mode observer; moreover, the torque threshold and the time threshold can be calibrated according to experiments, which can reduce the influence of interference torque on the detachment detection.

[0125] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting a hands-off steering wheel, characterized in that: The following steps are involved: Considering the inertia effect and damping effect of the steering wheel, a dynamic model of the vehicle steering system is established; Get the torsion bar torque and steering wheel angle: The driver's hand torque is observed by a driver's hand torque observer, wherein the driver's hand torque is used as a state variable, a state equation of the vehicle steering system is established in combination with a dynamic model of the vehicle steering system, and the driver's hand torque observer is established according to the state equation of the vehicle steering system and a super-helical sliding mode control algorithm; The torque threshold and time threshold are compared with the observed driver hand torque to obtain the hands-off detection result.

2. A method for detecting hands-off steering wheel according to claim 1, characterized in that: Taking the column power steering system as the research object, the dynamic models of the steering column and torque sensor are established respectively: Steering column: Torque sensor: K(θ sw -θ e )=T tb Among them, J sw is the moment of inertia of the steering wheel, B sw is the steering wheel damping coefficient, K is the torsion bar stiffness, θ sw is the steering wheel angle, θ e is the torsion bar output shaft angle, T d is the driver's hand torque, T tb is the torsion bar torque, represents the inertial torque caused by the inertial effect of the steering wheel, Represents the damping torque caused by the damping effect of the steering wheel.

3. The method for detecting hands-off steering wheel according to claim 1, characterized in that: In the acquisition of the torsion bar torque and the steering wheel angle, a combination of a TOS sensor + an angle encoder is used to respectively acquire the torsion bar torque and the power motor angle, and then the steering wheel angle is calculated by the transmission ratio between the motor angle and the steering wheel angle; or, the torsion bar torque and the steering wheel angle are acquired simultaneously by a TAS sensor.

4. The method for detecting hands-off steering wheel according to claim 1, characterized in that: The stability of the observer is verified according to Lyapunov's second method, and the range of values ​​of the observer parameters that meet the requirements is obtained to ensure that the observer can accurately observe the driver's hand torque.

5. The method for detecting hands-off steering wheel according to claim 1, characterized in that: The driver's hand torque T d is considered as a state variable, and The state equation of the vehicle steering system is established as: The driver hand torque observer based on the super-helical sliding mode algorithm is expressed as: Among them, U is the super-helical sliding mode control law, P is a function of U, representing the compensation amount of the steering wheel speed observation value, g is the feedback gain coefficient, is the estimated value of the steering wheel angle, ω is the angular velocity of the steering wheel, is the estimated value of the steering wheel angular velocity, is the estimated value of the driver's hand torque, is the derivative of the angular velocity of the steering wheel, is an estimate of the derivative of the steering wheel angle, T tb is the torsion bar torque, t is the time, J sw is the moment of inertia of the steering wheel, B sw is the steering wheel damping coefficient; Where the function P is: The super-helical sliding mode control law is expressed as: For the above formula, since J sw >0, in order to ensure that the estimated error of the driver's hand torque can quickly approach 0, let: Where k1 and k2 are design parameters and are greater than zero, and: Where s is the sliding surface, γ is a constant, and δ represents the upper limit of the absolute value of the bounded disturbance.

6. A method for detecting hands-off steering wheel according to any one of claims 1 to 5, characterized in that: The step of comparing the torque threshold and the time threshold with the observed driver's hand torque to obtain a hands-off detection result includes: If the observed driver's hand torque is less than the torque threshold, and the maintenance time of this state is greater than the time threshold, it is considered that the driver's hands have left the steering wheel, and this is a hands-off state; if the observed driver's hand torque is greater than the torque threshold, and the maintenance time of this state is greater than the time threshold, this is a non-hands-off state.

7. The method for detecting hands-off steering wheel according to claim 6, characterized in that: In each detection cycle, the observed driver's hand torque and torque threshold are compared. If the driver's hand torque is greater than the torque threshold, the counter value is added with a preset value. When the counter value reaches the maximum value, it means that the duration has reached the time threshold for holding judgment, and it is judged that the driver is holding the steering wheel with both hands; if the driver's hand torque is less than the torque threshold, the counter subtracts a preset value in each detection cycle. When the counter value reaches the minimum value, it means that the duration has reached the time threshold for hands-off judgment, and it is judged that the driver has both hands off the steering wheel. Otherwise, the judgment result is the same as the previous cycle.

8. A hands-off steering wheel detection system, characterized in that: For implementing the method according to any one of claims 1 to 7, the system comprises the following modules: Dynamic model building module, used to build the dynamic model of the vehicle steering system; Sensor signal acquisition module, used to obtain torsion bar torque and steering wheel angle: A driver hand torque observer observation module is used to observe the driver's hand torque through the driver's hand torque observer, wherein the driver's hand torque is used as a state variable, and a state equation of the vehicle steering system is established in combination with a dynamic model of the vehicle steering system, and the driver's hand torque observer is established according to the state equation of the vehicle steering system and a super-helical sliding mode control algorithm; The hands-off detection module is used to compare the torque threshold and the time threshold with the observed driver's hand torque to obtain a hands-off detection result.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for detecting hands-off steering wheel according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a method for detecting a hands-off steering wheel is implemented as described in any one of claims 1 to 7.

Images