A Vehicle Longitudinal Motion Control Method Based on Disturbance Estimation Compensation

Abstract

This invention relates to the field of vehicle control technology, specifically to a vehicle longitudinal motion control method based on disturbance estimation compensation. The method includes: calculating the speed error based on the planned speed from the upper-level planning system and the vehicle's actual speed; determining the desired acceleration based on the speed error and gradient compensation; determining the nominal acceleration based on the vehicle's actual speed, actual acceleration torque, and a nominal longitudinal acceleration model; determining the disturbance acceleration based on the actual acceleration and the nominal acceleration; determining the nominal coasting acceleration based on the vehicle's actual speed and the nominal longitudinal acceleration model; determining the coasting acceleration estimate based on the nominal coasting acceleration and the disturbance acceleration; and selecting acceleration drive or deceleration braking based on the coasting acceleration estimate and the desired acceleration value. This addresses the problem that existing control methods often fail to accurately represent the actual acceleration dynamics of a vehicle under different loads, wind resistances, and terrain conditions using the nominal longitudinal acceleration model.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and specifically to a method for longitudinal motion control of vehicles based on disturbance estimation compensation. Background Technology

[0002] Current longitudinal motion control methods for autonomous vehicles mainly include PID control, sliding mode control, and composite control methods consisting of speed closed-loop and acceleration open-loop control.

[0003] PID control is based on adjusting the control input to accelerate or decelerate the vehicle by feedback of speed error. PID control requires fine adjustment of control parameters. In addition, the control parameters will be different for different loads, wind resistances, and terrain conditions. Therefore, when PID control is applied to the longitudinal motion control system of a vehicle, the system's adaptability is not strong.

[0004] Sliding mode control is a method that designs a control law that enables the system state to quickly reach and maintain a predetermined sliding surface, thereby achieving accurate system control. In the application of vehicle longitudinal motion control systems, the sliding surface is usually used to keep the error between the set vehicle speed and the current vehicle speed zero. The control law quickly adjusts the acceleration drive or deceleration braking amount according to the vehicle speed error. Sliding mode control can cope with external disturbances to the system. However, the main problem with sliding mode control is that the high-frequency switching control can cause chattering, which can easily lead to system instability or reduced control performance.

[0005] The composite control method consisting of a speed closed-loop controller and an acceleration open-loop controller comprises a speed closed-loop controller and an acceleration open-loop controller. The speed closed-loop controller uses a PID controller to track the desired speed, eliminating the speed error between the actual vehicle speed and the target speed. The input to the controller is the speed deviation, and the output is the desired vehicle acceleration. The acceleration open-loop controller, based on the desired acceleration and the vehicle's current speed, calculates and outputs appropriate throttle or brake control values ​​by looking up a MAP table from the nominal longitudinal dynamics model calibration table, thereby controlling the vehicle to move at the desired acceleration. However, since the nominal longitudinal dynamics model calibration MAP table is constructed through offline testing, it reflects the vehicle's acceleration characteristics under specific operating conditions and cannot accurately describe the vehicle's acceleration characteristics under different loads, wind resistances, and terrain conditions, thus affecting the control system's adaptability under different operating conditions. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides a vehicle longitudinal motion control method based on disturbance estimation compensation.

[0007] The technical solution adopted by this invention to solve its technical problem is: a vehicle longitudinal motion control method based on disturbance estimation compensation, comprising the following steps:

[0008] S1: Receives planning speed information from the upper-level planning system, planning speed v exp Receive vehicle status information, including the actual vehicle speed v act Pitch information;

[0009] S2: Calculate the speed error based on the planned speed and the actual vehicle speed; the speed error v e =v exp -v act ;

[0010] S3: Calculate the slope compensation based on the vehicle's pitch angle. The slope θ of the road the vehicle is traveling on is equal to the vehicle's current pitch angle. The slope compensation is g*sinθ, where g is the acceleration due to gravity. Perform a first-order low-pass digital filter on the slope value θ: θ=a*pitch+(1-a)*θ l Where a is a coefficient of a first-order low-pass digital filter, θ l The slope value obtained from the previous calculation cycle;

[0011] S4: Calculate the desired acceleration a based on the speed error and the ramp compensation. exp :

[0012] a exp =k p *v e +g*sinθ;

[0013] If a exp >a exp When _max, a exp =a exp _max; such as a exp <-a exp When _max, a exp =-a exp _max, where k p v is the proportional coefficient of the proportional controller. e For speed error, g*sinθ is the ramp compensation amount, and a exp _max is the preset desired acceleration limit value;

[0014] S5: Based on vehicle speed v act and torque value T act Find the corresponding nominal acceleration 'a' in the nominal longitudinal acceleration model MAP table. nom If the vehicle speed or torque value does not perfectly match the data in the MAP table when looking it up, linear interpolation will be used for calculation.

[0015] S6: Based on vehicle speed v act Calculate the actual acceleration a of the vehicle act :

[0016] a act =(v act -v act _l) / dt;

[0017] a act =b*a act +(1-b)*a act _l;

[0018] Where v act _l represents the vehicle speed obtained in the previous calculation cycle, dt represents the sampling time, b represents the coefficients of the first-order low-pass digital filter, and a act _l represents the vehicle acceleration calculated in the previous calculation cycle;

[0019] S7: Based on the vehicle's actual acceleration a act and nominal acceleration a nom Calculate the disturbance acceleration a dis :

[0020] a dis =a act -a nom ;

[0021] S8: Based on vehicle speed v act Find the corresponding nominal gliding acceleration a in the nominal longitudinal acceleration model MAP table. nom_s If the vehicle speed does not perfectly match the data in the MAP table when looking it up, linear interpolation is used for calculation.

[0022] S9: Based on the estimated gliding acceleration value a sta and the expected acceleration value a exp The magnitude relationship between them determines whether to select acceleration drive control or deceleration braking control. If the acceleration value a exp Greater than the estimated gliding acceleration a sta +a z Then select acceleration drive control if the desired acceleration value a exp Less than the estimated gliding acceleration a sta -a z If the desired acceleration value is a, then deceleration braking control is selected. exp Greater than or equal to the estimated gliding acceleration a sta -a z And less than or equal to the estimated gliding acceleration a sta +a z The acceleration drive control torque output value and the deceleration control deceleration output value are both 0.

[0023] Preferably, the linear interpolation method of S5 is calculated as follows:

[0024]

[0025] Where T act T1 and T2 are the vehicle torque values ​​in the MAP table. act The endpoints of the torque range, V1 and V2, are the values ​​in the MAP table. act The endpoint of the velocity range, a 00 a 01 a 10 a 11 The nominal acceleration values ​​corresponding to (V1, T1), (V1, T2), (V2, T1), and (V2, T2) in the MAP table are given, where temp_acc1 and temp_acc2 are process variables, and a nom For (v act T act The corresponding nominal acceleration value.

[0026] Preferably, linear interpolation is used in step S8 for calculation:

[0027]

[0028] Where v act V1 and V2 are the vehicle speeds in the MAP table. act The endpoints of the velocity range, a0 and a1 are the nominal gliding acceleration values ​​corresponding to (V1, 0) and (V2, 0) in the MAP table, a nom_s To obtain the nominal gliding acceleration value.

[0029] Preferably, in step S9, deceleration braking control is selected to set the desired acceleration value a. exp The control output of the brake deceleration actuator controls the vehicle's braking and deceleration, while the acceleration drive control torque output value is 0.

[0030] Preferably, in step S9, acceleration drive control is selected based on the speed error v. e Calculate the acceleration torque compensation amount T com :

[0031]

[0032] Where K p K is the proportional coefficient of the proportional-integral (PI) controller. I The integral coefficient of the proportional-integral (PI) controller is dt, where dt is the sampling time; based on the desired acceleration a exp and disturbance acceleration a dis Calculate the required acceleration a dem :

[0033] a dem =aexp -a dis

[0034] Acceleration a according to demand dem and vehicle speed v act Find the corresponding required torque value T in the nominal longitudinal acceleration model MAP table. dem If the vehicle speed or acceleration value does not completely match the data in the MAP table when looking it up, linear interpolation is used for calculation.

[0035] Preferably, the linear interpolation method is calculated as follows:

[0036]

[0037] Where V1 and V2 are the vehicle speeds v in the MAP table. act The endpoints of the velocity range, a2(i) is the acceleration value corresponding to (V2, T(i)) in the MAP table, a1(i) is the acceleration value corresponding to (V1, T(i)) in the MAP table, and temp_acc(i) is the calculated acceleration value (v act The acceleration value corresponding to T(i) is i = 0, 1, 2…10, and a1 and a2 are the required acceleration values ​​in temp_acc(i). dem The endpoints of the interval, T1 and T2 are the torque values ​​corresponding to a1 and a2, respectively, T dem This is the final required torque value.

[0038] Preferably, the required torque value T dem and acceleration torque compensation amount T com The desired acceleration torque value T is obtained through calculation. exp :

[0039] T exp =T dem +T com ;

[0040] Desired acceleration torque value T exp The control output of the drive acceleration actuator controls the vehicle's driving acceleration, while the braking deceleration control output value is 0.

[0041] This invention has the following beneficial effects: By analyzing the model error between the nominal longitudinal acceleration model and the actual vehicle acceleration dynamics, this invention treats the model error as a system disturbance term. By estimating the system disturbance online and compensating it into the control system, it solves the problem that the nominal longitudinal acceleration model in existing control methods cannot accurately represent the actual acceleration dynamics of the vehicle under different loads, wind resistances, and terrain conditions. The system disturbance obtained from the online estimation is compensated into the coasting acceleration estimation and demand acceleration calculation, improving the stability and dynamic response capability of the control system. The control accuracy of the control system is improved by increasing the acceleration torque compensation. Attached Figure Description

[0042] Figure 1 This is a block diagram of the control structure of the present invention.

[0043] Figure 2 This is a flowchart illustrating the control method of the present invention.

[0044] Figure 3 This is the nominal longitudinal acceleration model MAP table of the present invention.

[0045] Figure 4 This is a 3D illustration of the nominal longitudinal acceleration model MAP table of the present invention.

[0046] Figure 5 This is a diagram showing the actual control effect of the present invention (I).

[0047] Figure 6 This is a diagram showing the actual control effect of the present invention (II). Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0049] like Figure 1-2 As shown, a vehicle longitudinal motion control method based on disturbance estimation compensation includes the following steps:

[0050] 1) Calculate the speed error based on the planned speed issued by the upper-level planning system and the actual vehicle speed;

[0051] 2) Determine the desired acceleration based on the speed error and the slope compensation; determine the nominal acceleration based on the actual vehicle speed, actual acceleration torque, and the nominal longitudinal acceleration model;

[0052] 3) Determine the disturbance acceleration based on the actual acceleration and the nominal acceleration; determine the nominal coasting acceleration based on the vehicle's actual speed and the nominal longitudinal acceleration model;

[0053] 4) Determine the estimated value of the coasting acceleration based on the nominal coasting acceleration and the disturbance acceleration; determine whether to use acceleration drive or deceleration braking based on the estimated value of the coasting acceleration and the expected acceleration value;

[0054] 5) If deceleration braking is selected, the desired acceleration value will be output to the vehicle's deceleration actuator to control the vehicle's deceleration braking.

[0055] 6) If acceleration drive is selected, the desired acceleration torque value is determined based on the desired acceleration, disturbance acceleration, inverse nominal longitudinal acceleration model and acceleration torque compensation amount, and the desired acceleration torque value is output to the vehicle's acceleration actuator to execute the vehicle acceleration drive.

[0056] The slope compensation is the component of gravitational acceleration on a slope when a vehicle is traveling on it. The slope compensation is calculated as g*sinθ, where g is the gravitational acceleration (approximately 9.81 m / s²). 2 ), where θ is the slope of the current road.

[0057] The desired acceleration is a combination of the velocity error after passing through the proportional controller and the ramp compensation, where the velocity error v is... e =v exp -v act v exp To plan the desired speed, v act The vehicle's current actual speed is given by the desired acceleration a. exp =k p *v e +g*sinθ,k p This is the proportional coefficient of the proportional controller.

[0058] Nominal acceleration is obtained by inputting the vehicle's actual speed and actual acceleration torque into a nominal longitudinal acceleration model. The nominal longitudinal acceleration model describes the mapping relationship between vehicle speed, acceleration, and torque under specific operating conditions. It represents the vehicle's acceleration response at different speeds and acceleration torques under specific conditions and is typically obtained by collecting and processing relevant data through real-vehicle testing. The nominal longitudinal acceleration model is usually represented by a MAP table, such as... Figure 3 As shown, the nominal longitudinal acceleration model (MAP) table is a two-dimensional table. The horizontal axis represents torque, and the vertical axis represents vehicle speed. Each intersection point corresponds to a specific acceleration value, based on the known vehicle speed v. act and torque T act Search in the nominal longitudinal acceleration model MAP table (v act T act The intersection of the two points corresponds to the nominal acceleration value a. nom .

[0059] The disturbance acceleration is a combination of the vehicle's actual acceleration and nominal acceleration, used to represent the error between the nominal longitudinal acceleration model and the longitudinal acceleration characteristics of the vehicle under actual operating conditions. The disturbance acceleration 'a' is... dis =a act -a nom a act a represents the actual acceleration of the vehicle. nom The nominal acceleration is obtained from the nominal longitudinal acceleration model MAP table.

[0060] Nominal coasting acceleration describes the relationship between vehicle speed and acceleration under specific operating conditions when the vehicle's driving torque is zero. When the vehicle's driving torque is 0, based on the known vehicle speed v... act Search in the nominal longitudinal acceleration model MAP table (v act The intersection point (0) corresponds to the nominal gliding acceleration value a. nom_s .

[0061] The coasting acceleration estimate is composed of the disturbance acceleration and the nominal coasting acceleration, and is used to represent the coasting acceleration value at different vehicle speeds under actual vehicle operating conditions. The coasting acceleration estimate is 'a'. sta =a dis +a nom_s .

[0062] The acceleration drive or deceleration braking switching control is based on the desired acceleration a. exp and the estimated gliding acceleration a sta The magnitude relationship between them is determined when the desired acceleration a exp Greater than the estimated gliding acceleration a sta When acceleration drive control is selected, and the desired acceleration a is... exp Less than the estimated gliding acceleration a sta When this is the case, select deceleration and braking control.

[0063] During deceleration and braking control, the desired acceleration value is used as the control quantity executed by the brake deceleration actuator, thereby controlling the vehicle's braking and deceleration.

[0064] During acceleration drive control, the desired acceleration torque is calculated based on the acceleration torque compensation, desired acceleration, disturbance acceleration, actual speed, and the inverse nominal longitudinal acceleration model. This desired acceleration torque value is then used as the control variable executed by the drive acceleration actuator, thereby controlling the vehicle's acceleration. The acceleration torque compensation is obtained by processing the speed error through a proportional-integral (PI) controller. K p K is the proportional gain of the PI controller. I These are the integral coefficients of the PI controller. The inverse nominal longitudinal acceleration model is the inverse of the nominal longitudinal acceleration model, i.e., it is derived from the current vehicle speed v. act and demand acceleration adem Search in the nominal longitudinal acceleration model MAP table (v act a dem The column containing the value is the required torque value T. dem Demand acceleration is composed of a combination of desired acceleration and disturbance acceleration, and demand acceleration a dem =a exp -a dis Ultimately, the desired acceleration torque value T exp =T dem +T com .

[0065] Example

[0066] A vehicle longitudinal motion control method based on disturbance estimation compensation includes the following steps:

[0067] 1) Receive planning speed information from the upper-level planning system, planning speed v exp .

[0068] 2) Receive vehicle status information, including the actual vehicle speed v act Pitch information.

[0069] 3) Calculate the speed error based on the planned speed and the actual vehicle speed. The speed error v e =v exp -v act , where v exp For planning speed, v act This refers to the actual vehicle speed.

[0070] 4) Calculate the slope compensation based on the vehicle's pitch angle. The slope θ of the road the vehicle is traveling on is equal to the vehicle's current pitch angle value. The slope compensation is g*sinθ, where g is the acceleration due to gravity (approximately 9.81 m / s²). 2 To suppress the impact of pitch angle jumps, a first-order low-pass digital filter is applied to the slope value θ: θ = a * pitch + (1 - a) * θ l Where a is the coefficient of the first-order low-pass digital filter, pitch is the current pitch angle of the vehicle, and θ l This is the slope value obtained from the previous calculation cycle.

[0071] 5) Calculate the desired acceleration a based on the speed error and the gradient compensation. exp The details are as follows:

[0072] a exp =k p *v e +g*sinθ;

[0073] For example, aexp >a exp When _max, a exp =a exp _max; such as a exp <-a exp When _max, a exp =-a exp _max.

[0074] Where k p v is the proportional coefficient of the proportional controller. e For speed error, g*sinθ is the ramp compensation amount, and a exp _max is the preset desired acceleration limit value.

[0075] 6) Based on the vehicle speed v act and torque value T act Find the corresponding nominal acceleration 'a' in the nominal longitudinal acceleration model MAP table. nom The nominal longitudinal acceleration model MAP table is as follows: Figure 3 As shown. Suppose we need to find the vehicle speed v. act =8m / s and torque value T act =50% acceleration, find the column corresponding to 50% torque value in the row with vehicle speed of 8 m / s. According to the MAP table, the intersection of vehicle speed of 8 m / s and torque value of 50% is 0.73 m / s. 2 That is, the nominal acceleration a at this time nom =0.73m / s 2 If the vehicle speed or torque value does not perfectly match the data in the MAP table when looking it up, linear interpolation can be used for calculation, as follows:

[0076]

[0077] Where v act For vehicle speed, T act T1 and T2 are the vehicle torque values ​​in the MAP table. act The endpoints of the torque range, V1 and V2, are the values ​​in the MAP table. act The endpoint of the velocity range, a 00 a 01 a 10 a 11 The nominal acceleration values ​​corresponding to (V1, T1), (V1, T2), (V2, T1), and (V2, T2) in the MAP table are given, where temp_acc1 and temp_acc2 are process variables, and a nom For (v act T act The corresponding nominal acceleration value.

[0078] 7) Based on the vehicle speed vact Calculate the actual acceleration a of the vehicle act The details are as follows:

[0079] a act =(v act -v act _l) / dt;

[0080] a act =b*a act +(1-b)*a act _l;

[0081] Where v act _l represents the vehicle speed obtained in the previous calculation cycle, dt represents the sampling time, b represents the coefficients of the first-order low-pass digital filter, and a act _l represents the vehicle acceleration calculated in the previous calculation cycle.

[0082] 8) Based on the vehicle's actual acceleration a act and nominal acceleration a nom Calculate the disturbance acceleration a dis The details are as follows:

[0083] a dis =a act -a nom ;

[0084] 9) Based on the vehicle speed v act Find the corresponding nominal gliding acceleration a in the nominal longitudinal acceleration model MAP table. nom_s The nominal longitudinal acceleration model MAP table is as follows: Figure 3 As shown, nominal coasting acceleration is the vehicle's acceleration at the current speed when torque is 0. Suppose we need to find the vehicle speed v. act The nominal coasting acceleration corresponding to 12 m / s² is found in the column corresponding to a torque value of 0 in the row for vehicle speed of 12 m / s². According to the MAP table, the intersection of vehicle speed of 12 m / s² and torque value of 0 is -0.4 m / s². 2 That is, the nominal gliding acceleration a nom_s = -0.4m / s 2 If the vehicle speed does not perfectly match the data in the MAP table when looking it up, linear interpolation can be used for calculation, as follows:

[0085]

[0086] Where v act V1 and V2 are the vehicle speeds in the MAP table. act The endpoints of the velocity range, a0 and a1 are the nominal gliding acceleration values ​​corresponding to (V1, 0) and (V2, 0) in the MAP table, a nom_sTo obtain the nominal gliding acceleration value.

[0087] 10) Based on the disturbance acceleration a dis and nominal gliding acceleration a nom_s Calculate the estimated value of gliding acceleration a sta The details are as follows:

[0088] a sta =a dis +a nom_s ;

[0089] 11) Based on the estimated gliding acceleration value a sta and the expected acceleration value a exp The relationship between the magnitudes determines whether to select acceleration drive control or deceleration braking control. To avoid frequent switching or high-frequency fluctuations between acceleration drive control and deceleration braking control during system adjustment, a dead-zone acceleration and deceleration switching logic is designed, such as the desired acceleration value 'a'. exp Greater than the estimated gliding acceleration a sta +a z Then select acceleration drive control, such as the desired acceleration value a. exp Less than the estimated gliding acceleration a sta -a z If the desired acceleration value is a, then deceleration and braking control is selected. exp Greater than or equal to the estimated gliding acceleration a sta -a z And less than or equal to the estimated gliding acceleration a sta +a z The acceleration drive control torque output value and the deceleration brake control deceleration output value are both 0.

[0090] If deceleration and braking control is selected, the desired acceleration value a will be... exp The control output of the brake deceleration actuator controls the vehicle's braking and deceleration, while the acceleration drive control torque output value is 0.

[0091] If acceleration drive control is selected, based on the speed error v e Calculate the acceleration torque compensation amount T com The details are as follows:

[0092]

[0093] Where K p K is the proportional coefficient of the proportional-integral (PI) controller. I dt represents the integral coefficient of the proportional-integral (PI) controller, and dt represents the sampling time.

[0094] 12) Based on the desired acceleration a exp and disturbance acceleration adis Calculate the required acceleration a dem The details are as follows:

[0095] a dem =a exp -a dis ;

[0096] 13) Acceleration a according to demand dem and vehicle speed v act Find the corresponding required torque value T in the nominal longitudinal acceleration model MAP table. dem Suppose we need to find the vehicle speed v act =8m / s and required acceleration a dem =0.88m / s 2 The required torque value, at a vehicle speed of 8 m / s, is found to correspond to a required acceleration of 0.88 m / s². 2 According to the MAP table, the corresponding columns show a vehicle speed of 8 m / s and an acceleration of 0.88 m / s². 2 The column corresponding to the intersection is 60%, which is the required torque value T. dem =60%. If the vehicle speed or acceleration value does not perfectly match the data in the MAP table when looking it up, linear interpolation can be used for calculation, as follows:

[0097]

[0098] Where V1 and V2 are the vehicle speeds v in the MAP table. act The endpoints of the velocity range, a2(i) is the acceleration value corresponding to (V2, T(i)) in the MAP table, a1(i) is the acceleration value corresponding to (V1, T(i)) in the MAP table, and temp_acc(i) is the calculated acceleration value (v act The acceleration value corresponding to T(i) is i = 0, 1, 2…10, and a1 and a2 are the required acceleration values ​​in temp_acc(i). dem The endpoints of the interval, T1 and T2 are the torque values ​​corresponding to a1 and a2, respectively, T dem This is the final required torque value.

[0099] 14) Based on the required torque value T dem and acceleration torque compensation amount T com Calculate the desired acceleration torque value T exp The details are as follows:

[0100] T exp =T dem +T com ;

[0101] Desired acceleration torque value T expThe control output of the drive acceleration actuator controls the vehicle's driving acceleration, while the braking deceleration control output value is 0.

[0102] Figure 5 The curve represents the response process of the vehicle's actual speed tracking the planned speed. The planned speed is set sequentially to 10 km / h, 20 km / h, 40 km / h, and 60 km / h, and then decelerated to 10 km / h. The vehicle's actual speed accelerates or decelerates sequentially to the set planned speed. Figure 6 The table shows the maximum deviation between the planned speed and the actual vehicle speed at different speeds, such as... Figure 6 As shown in (a), when the planned speed is set at 10 km / h, the maximum speed error is 0.45 km / h. Figure 6 As shown in (b), when the planned speed is set at 20 km / h, the maximum speed error is 0.30 km / h. Figure 6 As shown in (c), when the planned speed is set at 40 km / h, the maximum speed error is 0.55 km / h. Figure 6 As shown in (d), when the planned speed is set to 60 km / h, the maximum speed error is 0.96 km / h. Analysis of the actual control effect verifies that this method can meet the requirements of the upper-level system for longitudinal motion control of the vehicle.

[0103] Definitions of abbreviations and key terms: Autonomous vehicle motion control primarily involves controlling the vehicle platform to track the real-time planned trajectory (path, speed) output by the upper-level planning system. Vehicle motion control is mainly divided into lateral motion control and longitudinal motion control. Lateral motion control achieves accurate path tracking by controlling the vehicle's steering system. Lateral motion control calculates the desired steering information based on the real-time planned path output by the upper-level planning system and the vehicle's own state information, controlling the vehicle to travel along the planned path. Longitudinal motion control, based on the speed command output by the upper-level planning system and the vehicle's own state information, achieves precise tracking of the desired speed by coordinating the torque of the vehicle's powertrain and the braking amount of the braking system.

[0104] The nominal longitudinal acceleration model (MAM) describes the mapping relationship between vehicle speed, acceleration, and torque under specific operating conditions. It represents the vehicle's acceleration response at different speeds and acceleration torques under those conditions. The MAM is typically represented by a MAP table, a two-dimensional table where the horizontal axis represents torque and the vertical axis represents vehicle speed. Each intersection point corresponds to a specific acceleration value. MAM tables are usually obtained by collecting and processing data from real-vehicle tests. The nominal acceleration is calculated by interpolation after inputting the actual vehicle speed and actual acceleration torque into the MAM table. Disturbance estimation is used to analyze the relationship between the nominal longitudinal acceleration model and the actual vehicle acceleration dynamics characteristics and to estimate system disturbances; it is the difference between the actual and nominal acceleration values.

[0105] This invention is not limited to the above-described embodiments. Anyone should know that any structural changes made under the guidance of this invention, and any technical solutions that are the same as or similar to this invention, fall within the protection scope of this invention.

[0106] The technologies, shapes, and structures not described in detail in this invention are all known technologies.

Claims

1. A vehicle longitudinal motion control method based on disturbance estimation compensation, characterized in that, Includes the following steps: S1: Receives planning speed information from the upper-level planning system, planning speed v exp Receive vehicle status information, including the actual vehicle speed v act Pitch information; S2: Calculate the speed error based on the planned speed and the actual vehicle speed; the speed error v e =v exp -v act ; S3: Calculate the slope compensation based on the vehicle's pitch angle. The slope θ of the road the vehicle is traveling on is equal to the vehicle's current pitch angle. The slope compensation is g*sinθ, where g is the acceleration due to gravity. Perform a first-order low-pass digital filter on the slope value θ: θ=a*pitch+(1-a)*θ l Where a is a coefficient of a first-order low-pass digital filter, θ l The slope value obtained from the previous calculation cycle; S4: Calculate the desired acceleration a based on the speed error and the ramp compensation. exp : a exp =k p *v e +g*sinθ; If a exp >a exp When _max, a exp =a exp _max; such as a exp <-a exp When _max, a exp =-a exp _max, where k p v is the proportional coefficient of the proportional controller. e For speed error, g*sinθ is the ramp compensation amount, and a exp _max is the preset desired acceleration limit value; S5: Based on vehicle speed v act and torque value T act Find the corresponding nominal acceleration 'a' in the nominal longitudinal acceleration model MAP table. nom If the vehicle speed or torque value does not perfectly match the data in the MAP table when looking it up, linear interpolation will be used for calculation. S6: Based on vehicle speed v act Calculate the actual acceleration a of the vehicle act : and act =(in act -in act _l) / dt; a act =b*a act +(1-b)*a act _l; Where v act _l represents the vehicle speed obtained in the previous calculation cycle, dt represents the sampling time, b represents the coefficients of the first-order low-pass digital filter, and a act _l represents the vehicle acceleration calculated in the previous calculation cycle; S7: Based on the vehicle's actual acceleration a act and nominal acceleration a nom Calculate the disturbance acceleration a dis : a dis =a act -a nom ; S8: Based on vehicle speed v act Find the corresponding nominal gliding acceleration a in the nominal longitudinal acceleration model MAP table. nom_s If the vehicle speed does not perfectly match the data in the MAP table when looking it up, linear interpolation is used for calculation. S9: Based on the estimated gliding acceleration value a sta and the expected acceleration value a exp The magnitude relationship between them determines whether to select acceleration drive control or deceleration braking control. If the acceleration value a exp Greater than the estimated gliding acceleration a sta +a z Then select acceleration drive control if the desired acceleration value a exp Less than the estimated gliding acceleration a sta -a z If the desired acceleration value is a, then deceleration braking control is selected. exp Greater than or equal to the estimated gliding acceleration a sta -a z And less than or equal to the estimated gliding acceleration a sta +a z The acceleration drive control torque output value and the deceleration control deceleration output value are both 0.

2. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 1, characterized in that, The linear interpolation method of S5 is calculated as follows: Where T act T1 and T2 are the vehicle torque values ​​in the MAP table. act The endpoints of the torque range, V1 and V2, are the values ​​in the MAP table. act The endpoint of the velocity range, a 00 a 01 a 10 a 11 The nominal acceleration values ​​corresponding to (V1, T1), (V1, T2), (V2, T1), and (V2, T2) in the MAP table are given, where temp_acc1 and temp_acc2 are process variables, and a nom For (v act T act The corresponding nominal acceleration value.

3. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 1, characterized in that, The S8 calculation uses linear interpolation: Where v act V1 and V2 are the vehicle speeds in the MAP table. act The endpoints of the velocity range, a0 and a1 are the nominal gliding acceleration values ​​corresponding to (V1, 0) and (V2, 0) in the MAP table, a nom_s To obtain the nominal gliding acceleration value.

4. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 1, characterized in that, In S9, deceleration and braking control is selected, and the desired acceleration value a is... exp The control output of the brake deceleration actuator controls the vehicle's braking and deceleration, while the acceleration drive control torque output value is 0.

5. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 1, characterized in that, In S9, acceleration drive control is selected based on the speed error v. e Calculate the acceleration torque compensation amount T com : Where K p K is the proportional coefficient of the proportional-integral (PI) controller. I The integral coefficient of the proportional-integral (PI) controller is dt, where dt is the sampling time; based on the desired acceleration a exp and disturbance acceleration a dis Calculate the required acceleration a dem : a dem =a exp -a dis Acceleration a according to demand dem and vehicle speed v act Find the corresponding required torque value T in the nominal longitudinal acceleration model MAP table. dem If the vehicle speed or acceleration value does not completely match the data in the MAP table when looking it up, linear interpolation is used for calculation.

6. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 5, characterized in that, The linear interpolation method is calculated as follows: Where V1 and V2 are the vehicle speeds v in the MAP table. act The endpoints of the velocity range, a2(i) is the acceleration value corresponding to (V2, T(i)) in the MAP table, a1(i) is the acceleration value corresponding to (V1, T(i)) in the MAP table, and temp_acc(i) is the calculated acceleration value (v act The acceleration value corresponding to T(i) is i = 0, 1, 2…10, and a1 and a2 are the required acceleration values ​​in temp_acc(i). dem The endpoints of the interval, T1 and T2 are the torque values ​​corresponding to a1 and a2, respectively, T dem This is the final required torque value.

7. The vehicle longitudinal motion control method based on disturbance estimation compensation according to claim 5, characterized in that, The required torque value T dem and acceleration torque compensation amount T com The desired acceleration torque value T is obtained through calculation. exp : T exp =T dem +T com ; Desired acceleration torque value T exp The control output of the drive acceleration actuator controls the vehicle's driving acceleration, while the braking deceleration control output value is 0.

Images