Adaptive compensation algorithm for commercial vehicle power steering based on model predictive control

Abstract

The model prediction control-based mid-position adaptive compensation algorithm for commercial vehicle power steering comprises the following steps: acquiring vehicle running parameters in real time, starting the mid-position adaptive compensation mode when the vehicle running parameters meet the set conditions, calculating the optimal front wheel steering angle δ k based on the Kalman estimation method, establishing a model prediction control problem according to a vehicle EPS system model and a mid-position adaptive compensation target, and obtaining the optimal control variable ΔT com,i at each moment, which is applied to the mid-position compensation torque change of the EPS until the driver's steering hand force gradually decreases to 0, thereby realizing the mid-position adaptive compensation of the vehicle power steering. The present application identifies the front wheel steering angle through the Kalman estimation algorithm and adopts the model prediction control method to adaptively compensate the driver's steering hand force, thereby reducing the driver's steering hand force, preventing the vehicle from deviating, reducing the driving fatigue of the driver caused by long-term hand steering, and ensuring the driving safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile steering control, in particular to a commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control. BACKGROUND

[0002] As a main control object of integrated control system, electric power steering has an important influence on the driving comfort and the steering stability of the automobile during steering. The electric power steering system (EPS) can provide appropriate steering assistance in real time according to the steering demand of the driver and the driving state of the automobile, and can also better solve the contradiction between the light steering requirement at low speed and the stable steering requirement at high speed. Compared with the mechanical steering system and the hydraulic power steering system, the EPS also has the advantages of safety, environmental protection, energy saving and simple assembly. With the continuous development and improvement of the control method, the function of the EPS is no longer limited to steering assistance, but also can realize active return, road disturbance suppression, lane keeping, lane deviation warning and automatic parking and other functions.

[0003] During the use of the commercial vehicle, as the mechanical wear in the steering machine gradually increases, the mid-position of the steering wheel will be offset. If the controller ECU still refers to the unchanged intermediate position, it will cause errors in the calculated steering wheel angle, and then affect the calculation of the assist torque, causing the vehicle to deviate.

[0004] The deviation of the automobile driving is that the driver needs to adjust the steering wheel to the intermediate position to keep the vehicle straight on the longitudinal axis. This requires the driver to exert additional corrective force to maintain the straight driving of the vehicle, which not only increases the burden of single-sided steering, but also may cause the driver to be tired. The mid-position offset not only affects the comfort of driving, but also may cause abnormal tire wear, and in severe cases, it may even cause the vehicle to deviate, and even the car to lose control. SUMMARY

[0005] The present application provides a commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control, which mainly aims to solve the problem of mid-position offset of the steering wheel of the commercial vehicle in the prior art.

[0006] The present application adopts the following technical solutions:

[0007] The commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control comprises the following steps:

[0008] Step 1: Real-time acquisition of vehicle driving parameters, when the vehicle driving parameters meet the set conditions, the mid-position adaptive compensation mode is started;

[0009] Step 2: Calculate the optimal front wheel angle δ based on the Kalman estimation method k ;

[0010] Step 3: Establish the model predictive control problem according to the vehicle EPS system model and the median adaptive compensation target model, and solve the optimal control variable AT at each time com,i , which acts on the median compensation torque change of the EPS until the driver's steering hand force gradually decreases to 0, thereby realizing the median adaptive compensation of the vehicle power-assisted steering; the established model predictive control problem is:

[0011]

[0012] s.t.

[0013]

[0014] In the formula: J is the model predictive control target function, which is designed to reduce the driver's steering hand force; w1 and w2 represent weight coefficients; i = 1, …, N p represents the prediction time; δ k,i represents the optimal front wheel angle obtained by Kalman estimation at time i; δ f,i represents the vehicle front wheel angle when the steering hand force at time i is 0; ΔT com,i represents the median compensation torque change at time i; T com,i is the median compensation torque of the EPS at time i; J eq is the equivalent moment of inertia of the system; C eq is the equivalent damping of the system; is the angular velocity of the steering input shaft; is the angular acceleration of the steering input shaft at time i; T r,i is the simplified steering resistance torque at time i; T motor,i is the steering assist calculated according to the steering wheel torque at time i; δ P,i is the angle of the steering input shaft at time i; N is the transmission ratio of the steering input shaft to the front wheel angle at time i; ΔT com,min and ΔT com,max respectively represent the minimum and maximum values of the median compensation torque change; T com,min and T com,max respectively represent the minimum and maximum values of the median compensation torque.

[0015] Further, in step 1, the vehicle running parameters include vehicle speed, steering wheel angle, steering wheel speed, steering wheel torque, four-wheel speed, speed difference, and front and rear axle speed difference.

[0016] Furthermore, the setting conditions for turning on the EPS mid-position adaptive compensation mode are as follows: (1) the vehicle speed is greater than a preset threshold value S1; (2) the steering wheel angle is less than a preset threshold value S2; (3) the steering wheel speed is less than a preset threshold value S3; (4) the steering wheel torque is less than a preset threshold value S4; (5) the difference between the left and right wheel speeds and the vehicle speed is less than a preset threshold value S5; (6) the speed difference between the front and rear axles is less than a preset threshold value S6; (7) the duration of simultaneously satisfying conditions (1)-(6) exceeds a preset threshold value S7.

[0017] Furthermore, in step 2, the Ackerman steering kinematics model is first used to calculate the front wheel turning angle δ according to the vehicle front axle turning radius R0, and then the Kalman estimation method is used to optimize the front wheel turning angle δ to obtain the optimal front wheel turning angle δ k .

[0018] Furthermore, the established model predictive control problem is transformed into a quadratic programming problem and solved to obtain the optimal median compensation torque control sequence [ΔT com,1 ,ΔT com,2 ,…,ΔT com,Np ], and the first control variable of the optimal control sequence acts on the median compensation torque change of EPS.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention uses the Kalman estimation algorithm to identify the front wheel angle and adopts the model predictive control method to adaptively compensate the driver's steering force. As a result, when the car deviates, the EPS can accurately provide an additional steering assist compensation, thereby reducing the driver's steering force, preventing the car from deviating, and reducing the driver's driving fatigue caused by long-term hand-holding on the steering wheel, thereby ensuring driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Schematic diagram of the control flow of the present invention.

[0022] Figure 2 Schematic diagram of the Ackerman steering kinematic model.

[0023] Figure 3 This is the steering wheel median angle change diagram after median compensation. DETAILED DESCRIPTION

[0024] The specific embodiments of the present invention are described below with reference to the accompanying drawings. In order to fully understand the present invention, many details are described below, but for those skilled in the art, the present invention can be implemented without these details.

[0025] like Figure 1As shown, the main purpose of the present invention is to provide a mid-range adaptive compensation algorithm for commercial vehicle power steering based on model predictive control. This algorithm enables the electric power steering to provide an adaptive deviation compensation amount based on the deviation degree after the EPS (Electric Steering System) detects the vehicle's deviation. The EPS deviation compensation amount is iteratively increased as the deviation degree (driver's hand pressure on the steering wheel) increases, thereby achieving adaptive adjustment of the mid-range deviation compensation amount.

[0026] like Figure 1 To elaborate on the design concept of the present invention, the algorithm provided by the present invention is described in detail below by collecting driving data of an experimental vehicle. The specific implementation steps are as follows:

[0027] Step 1: Obtain vehicle driving parameters in real time. When the vehicle driving parameters meet the setting conditions of the EPS median adaptive compensation mode, the median adaptive compensation mode is turned on.

[0028] In this example, the steering wheel center position of the experimental vehicle was roughly calibrated to within approximately ±10° of the driving center position. The vehicle's structural parameters for information collection are as follows: a 2-meter wheelbase (W), a 5.7-meter wheelbase (L), and a steering wheel transmission ratio of 24.3. During the information collection process, the experimental vehicle underwent a straight-line acceleration and deceleration test, with data collection lasting 170 seconds.

[0029] Vehicle driving parameters include vehicle speed, steering wheel angle, steering wheel speed, steering wheel torque, four-wheel speed, vehicle speed difference, and front and rear axle speed difference. The setting conditions for turning on EPS mid-position adaptive compensation mode are:

[0030] (1) The vehicle speed is greater than a preset threshold value S1, where the preset value S1 is a calibration value, and in this embodiment, it is preferably 40 km / h;

[0031] (2) The steering wheel angle is less than a preset threshold value S2, where the preset value S2 is a calibration value, and in this embodiment, it is preferably 3°;

[0032] (3) The steering wheel speed is less than a preset threshold value S3, where the preset value S3 is a calibration value, preferably 4° / s in this embodiment;

[0033] (4) The steering wheel torque is less than a preset threshold value S4, where the preset value S4 is a calibration value, preferably 0.5 N·m in this embodiment;

[0034] (5) The difference between the left and right wheel speeds and the vehicle speed is less than a preset threshold value S5, where the preset value S5 is a calibration value, preferably 5% in this embodiment;

[0035] (6) The front and rear axle speed difference is less than a preset threshold value S6, where the preset value S6 is a calibration value, preferably 5% in this embodiment;

[0036] (7) The duration of simultaneously satisfying conditions (1)-(6) exceeds a preset threshold value S7, where the preset value S7 is a calibration value, and in this embodiment, it is preferably 5s.

[0037] When the vehicle meets conditions (1)-(7) at the same time, the EPS turns on the mid-position adaptive compensation mode.

[0038] Step 2: Calculate the optimal front wheel steering angle δ based on the Kalman estimation method k .

[0039] Specifically, this step first adopts Figure 2 The Ackermann steering kinematic model is shown, and the front wheel turning angle δ is calculated based on the vehicle's front axle turning radius R0:

[0040]

[0041] Where: δ L is the left front wheel turning angle, δ R is the right front wheel turning angle, V FL is the left front wheel speed, V FR is the speed of the right front wheel, W is the track width, L is the wheelbase, and R0 is the front axle turning radius.

[0042] In order to improve the calculation accuracy, the Kalman estimation method is used to optimize the front wheel steering angle δ, so as to obtain the optimal front wheel steering angle δ k , the first-order Kalman estimation equation is simplified as follows:

[0043]

[0044]

[0045]

[0046]

[0047]

[0048] Where: k represents the time scale; represents the predicted value of the front wheel angle at time k; Q and R represent the prediction error and measurement error of δ respectively; P k represents the system covariance; Represents the prediction system covariance value; K k represents the Kalman gain; z k represents the front wheel turning angle measured by the sensor; δ k represents the optimal front wheel steering angle obtained by Kalman estimation.

[0049] The reason why the present invention uses the Kalman estimation method to estimate the front wheel steering angle is that for the same specific angle, each estimation will have errors. If mean iteration is used, it can converge to a satisfactory result after a certain number of measurements. However, if the wear and deformation of the mechanical structure occurs during driving, causing the median to change, it will be difficult to converge to the new median within a small number of iterations. The Kalman estimation method can simultaneously accomplish two tasks: improving estimation accuracy through multiple measurements and quickly adapting to new measurement values. Therefore, this estimation method is used in the processing of time series data, thereby ensuring that it can adapt to the median offset problem caused by mechanical wear in the steering gear.

[0050] Step 3: Establish a model predictive control problem based on the vehicle EPS system model and the median adaptive compensation target, and find the optimal control variable ΔT at each moment com,i , and apply it to the median compensation torque change of EPS until the driver's steering hand force gradually decreases to 0, thereby realizing the vehicle power steering median adaptive compensation.

[0051] Specifically, the EPS median compensation torque change ΔT com As the control variable predicted by the model, the vehicle front wheel angle δ f As a state variable, the model predictive control objective function J is designed to reduce the driver's steering force, that is, in the prediction time domain N p When the driver's steering force is 0, the vehicle's front wheel angle δ f,i The front wheel steering angle δ obtained by Kalman estimation k,i Consistent, that is:

[0052]

[0053] Where: J is the model predictive control objective function, which is designed to reduce the driver's steering force; w1 and w2 are weight coefficients; i = 1, ..., N p represents the prediction time; δ k,i represents the optimal front wheel steering angle obtained by Kalman estimation at time i, which is calculated in step 2; δ f,i represents the front wheel angle of the vehicle when the steering force is 0 at time i; ΔT com,i Indicates the change in median compensation torque at time i.

[0054] Prediction of vehicle front wheel angle δ based on electric power steering system dynamics model f , and recorded as the vehicle front wheel angle sequence [δ f,1 ,δ f,2 ,…,δ f,NP ]:

[0055]

[0056] Where: δ P,i is the angle of the steering input shaft at time i; N G is the transmission ratio from the steering input shaft to the front wheel angle at time i.

[0057] Equivalent the inertia and damping of the entire system to the output end of the recirculating ball, the simplified one-degree-of-freedom model of the steering system is obtained:

[0058]

[0059] Where: J eq is the equivalent moment of inertia of the system; C eq is the equivalent damping of the system; is the angular velocity of the steering gear input shaft; is the angular acceleration of the steering input shaft at time i; T r,i is the simplified steering resistance torque at time i; T motor,i is the steering assist force calculated based on the steering wheel torque at time i; T com,i is the median compensation torque of EPS at time i, and its calculation formula is:

[0060] T com,i =T com,i-1 +ΔT com,i (3.3)

[0061] The optimal model predictive control problem established is as follows:

[0062]

[0063] st

[0064]

[0065] Where: ΔT com,min and ΔT com,max Respectively represent the minimum and maximum values ​​of the change in the median compensation torque; T com,min and T com,max Represent the minimum and maximum values ​​of the median compensation torque respectively.

[0066] The model predictive control problem established above, namely formulas (2) and (3), is transformed into a quadratic programming problem and solved to obtain the optimal median compensation torque control sequence [ΔT com,1 ,ΔT com,2 ,…,ΔT com,Np ], and the first control variable of the optimal control sequence acts on the median compensation torque change of EPS (that is, the median compensation torque change ΔT obtained at each moment com,1 Output corresponding mid-position compensation torque T com,i) until the driver's steering hand force gradually decreases to 0, thereby achieving vehicle power steering mid-position adaptive compensation.

[0067] The compensation results of this embodiment are as follows Figure 3 As shown. Figure 3 It can be seen that the entire segment of data triggered a total of 6 median adaptive compensations, that is, the algorithm identified 6 segments of approximate straight-line conditions, and each segment of approximate straight-line conditions triggered the compensation angle mechanism once, obtaining the corresponding median compensation torques for the 6 segments. At the same time, due to the torque compensation effect of the model predictive control algorithm on the median angle, the compensated median angle gradually converged to a position that was 4° different from the original initial angle within 6 measurements. It can be seen that the compensated median angle showed a stable trend after the gradual convergence process, which shows that the adaptive compensation algorithm can effectively calibrate the steering wheel median error. When the median angle compensation is stable, the deviation angle is consistent with the initial steering wheel median error range, which further verifies the effectiveness of the adaptive compensation algorithm provided by the present invention.

[0068] This invention uses a Kalman estimation algorithm to identify the front wheel angle and a model predictive control method to adaptively compensate for the driver's steering force. Consequently, if the vehicle veers, the EPS can accurately provide additional steering assistance, thereby reducing the driver's steering force, preventing the vehicle from veering off the track, and alleviating driver fatigue caused by prolonged hands-on steering, thereby ensuring driving safety. Compared to existing technologies, this invention has two main differences:

[0069] (1) Different from the prior art which realizes zero-position compensation of the steering angle sensor through self-learning under straight-ahead driving conditions, the present invention does not rely on the steering angle sensor, but is based on the wheel speed sensor information and utilizes the Kalman estimation algorithm to identify the front wheel steering angle. At the same time, the steering hand force zero-position is adaptively compensated by adjusting the power-assist torque, thereby achieving the purpose of steering angle zero-position compensation.

[0070] (2) Compared with low-pass filter compensation, the model predictive control method adopted in the present invention takes into account the steering system dynamics model and performs rolling optimization on the median compensation torque, which has the advantages of good control effect and strong robustness.

[0071] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited to this. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.

Claims

1. A commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control, characterized by: The steps include: Step 1: Obtain vehicle driving parameters in real time. When the vehicle driving parameters meet the set conditions, the mid-position adaptive compensation mode is turned on. Step 2: Calculate the optimal front wheel steering angle δ based on the Kalman estimation method k ; Step 3: Establish a model predictive control problem based on the vehicle EPS system model and the median adaptive compensation target, and find the optimal control variable ΔT at each moment com,i , which acts on the EPS's median compensation torque change until the driver's steering hand force gradually decreases to 0, thereby achieving the vehicle's power steering median adaptive compensation; the established model predictive control problem is: st Where: J is the model predictive control objective function, which is designed to reduce the driver's steering force; w1 and w2 represent weight coefficients; i=1,…,N p represents the prediction time; δ k,i represents the optimal front wheel steering angle obtained by Kalman estimation at time i; δ f,i represents the front wheel angle of the vehicle when the steering force is 0 at time i; ΔT com,i represents the change in the median compensation torque at time i; T com,i is the median compensation torque of EPS at time i; J eq is the equivalent moment of inertia of the system; C eq is the equivalent damping of the system; is the angular velocity of the steering gear input shaft; is the angular acceleration of the steering input shaft at time i; T r,i is the simplified steering resistance torque at time i; T motor,i is the steering assist force calculated based on the steering wheel torque at time i; δ P,i is the angle of the steering input shaft at time i; N G is the transmission ratio from the steering input shaft to the front wheel angle at time i; ΔT com,min and ΔT com,max Respectively represent the minimum and maximum values ​​of the change in the median compensation torque; T com,min and T com,max Represent the minimum and maximum values ​​of the median compensation torque respectively.

2. The commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control according to claim 1, characterized in that: In step 1, the vehicle driving parameters include vehicle speed, steering wheel angle, steering wheel speed, steering wheel torque, four-wheel speed, vehicle speed difference and front and rear axle speed difference.

3. The commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control according to claim 2, characterized in that: The setting conditions for turning on the EPS mid-position adaptive compensation mode are: (1) the vehicle speed is greater than the preset threshold value S1; (2) the steering wheel angle is less than the preset threshold value S2; (3) the steering wheel speed is less than the preset threshold value S3; (4) the steering wheel torque is less than the preset threshold value S4; (5) the difference between the left and right wheel speeds and the vehicle speed is less than the preset threshold value S5; (6) the speed difference between the front and rear axles is less than the preset threshold value S6; (7) the duration of simultaneously meeting conditions (1)-(6) exceeds the preset threshold value S7.

4. The commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control according to claim 1, characterized in that: In step 2, the Ackerman steering kinematic model is first used to calculate the front wheel steering angle δ according to the vehicle front axle turning radius R0, and then the Kalman estimation method is used to optimize the front wheel steering angle δ to obtain the optimal front wheel steering angle δ k .

5. The commercial vehicle power steering mid-position adaptive compensation algorithm based on model predictive control according to claim 1, characterized in that: The established model predictive control problem is transformed into a quadratic programming problem and solved to obtain the optimal median compensation torque control sequence [ΔT com,1 ,ΔT com,2 ,…,ΔT com,Np ], and the first control variable of the optimal control sequence acts on the median compensation torque change of EPS.

Images