RU2785446C1 - Method and device for control of lateral movement of self-drivering vehicle and self-driving vehicle - Google Patents

Abstract

FIELD: self-driving vehicles.

SUBSTANCE: invention relates to self-driving vehicles. In the method for controlling the lateral movement of a self-controlled vehicle, the control parameters of a linear-quadratic controller are set; obtaining the general TC parameters and the gain of the control parameter in accordance with the control parameters and the general parameters. The gain of the control parameter is obtained in real time in accordance with the gain of the control parameter. The value of the feedback on the state error, the curvature of the trajectory and the angle of inclination of the vehicle are obtained. The control and compensating values are obtained in accordance with the real-time control parameter gain, the status error feedback value, the trajectory curvature, and the vehicle inclination angle. The expected steering angle is obtained according to the control and compensation value, and the expected steering angle is output to the steering-by-wire system.

EFFECT: volume of calculations and computing resources is reduced.

10 cl, 8 dwg

Description

[0001] The present application claims priority from Chinese Patent Application No. 202010456541.0 filed with the National Intellectual Property Office of China on May 26, 2020 and entitled "Method and apparatus for controlling the lateral movement of a self-driving vehicle and a self-driving vehicle", the entire content of which is included in this document by reference.

Technical field

[0002] The present invention relates to the technical field of self-driving vehicles, in particular, to a method and apparatus for controlling the lateral movement of a self-driving vehicle, and to a self-driving vehicle.

State of the art

[0003] The lateral motion control system of the self-driving vehicle calculates a desired steering command in real time according to the planned trajectory and upper level information about the position of the vehicle, and controls the vehicle to move along the target trajectory. At present, the optimal control parameter gain is usually obtained by solving the Riccati equation online, which is computationally intensive and requires more computing resources.

Disclosure of the essence of the invention

[0004] The present invention is directed to solving or correcting at least one of the above technical problems.

[0005] To this end, the first object of the present invention is to provide a method for controlling the lateral movement of a self-driving vehicle.

[0006] The second object of the present invention is to provide a device for controlling the lateral movement of a self-driving vehicle.

[0007] The third object of the present invention is to provide a self-driving vehicle.

[0008] To solve the first object of the present invention, embodiments of the present invention provide a method for controlling the lateral movement of a self-driving vehicle, including: setting control parameters of a linear-quadratic controller; getting the general parameters of the vehicle; obtaining a gain of the control parameter in accordance with the control parameters and the general parameters of the vehicle; obtaining a gain of the control parameter in real time in accordance with the gain of the control parameter; obtaining a feedback value on the status error, the curvature of the trajectory and the angle of the vehicle; obtaining a control quantity and a compensation quantity according to a real-time control parameter gain, a state error feedback quantity, a path curvature, and a vehicle tilt angle; and obtaining an expected steering angle according to the control variable and the compensation variable, and outputting the expected steering angle to the steering-by-wire system.

[0009] This embodiment makes it possible to realize the optimal control of the lateral movement of a self-driving vehicle without static error, significantly reduce the amount of calculations, ensure the convergence of the optimal gain of the control system, and take into account the influence of road curvature, road slope and uncertainty of the overall parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0010] In addition, the embodiments provided by the above embodiments of the present invention may also have the following additional technical features.

[0011] In the above embodiment, the step of setting the control parameters of the linear-quadratic controller includes: the control parameters include the first weight matrix Q and the second weight matrix R, Q=diag[q ₁ , q ₂ , q ₃ , q ₄ ], R=[ r], where q ₁ , q ₂ , q ₃ , q ₄ and r are the five control parameters, respectively, q2, q4 and r are set to fixed values, and q1, q3 are obtained according to the following formula:

[0012] Where q1max is the maximum value of q1, q1min is the minimum value of q1, q3max is the maximum value of q3, q3min is the minimum value of q3, Vx is the longitudinal speed of the vehicle, V1 is the first speed threshold, and V2 is the second speed threshold.

[0013] In this embodiment, q ₂ , q ₄ , and r are set to fixed values, the parameters q ₁ , q ₃ are linearly adjusted in accordance with the longitudinal speed V _x of the vehicle, that is, the weight matrix of the control parameters of the linear-quadratic controller uses speed linear adjustment method, which can greatly reduce the amount of computation.

[0014] In any of the above embodiments, the step of obtaining the control parameter gain according to the control parameters and the general vehicle parameters includes: obtaining the control parameter gain according to the following formula:

[0015] Where m is the mass of the vehicle, I _z is the inertia of rotation of the vehicle about the vertical direction, I _f is the distance from the front axle of the vehicle to the center of mass, I _r is the distance from the rear axle of the vehicle to the center of mass, C _f is front wheel stiffness in cornering and C _r rear wheel stiffness in cornering, V _x longitudinal vehicle speed, k ₁₄ convergence matrix of control gain, k ₁₄₍₁₎ , k ₁₄₍₂₎ , k _{14( 3)} and k ₁₄₍₄₎ are four elements in the convergence matrix of the control system gain, respectively, k ₁ , k ₂ , k ₃ and k ₄ are four control parameter gains, respectively.

[0016] General parameters of the vehicle include the mass of the vehicle and/or the inertia of rotation of the vehicle about the vertical direction and/or the distance from the front axle of the vehicle to the center of mass and/or the distance from the rear axle of the vehicle to the center of mass and/or the stiffness of the front wheels when cornering and/or stiffness of the rear wheels when cornering. In accordance with the general parameters of the vehicle and the model of the controlled object, the optimal coefficients k ₁ , k ₂ , k ₃ and k ₄ of the control parameter gain are calculated offline using the Iqr function in MATLAB, which significantly reduces the amount of calculation compared to the parameter gain factor solution online control.

[0017] In any of the above embodiments, the step of obtaining a real-time control parameter gain in accordance with the control parameter gain includes: taking the speed as an independent variable and performing a polynomial approximation of the control parameter gain to obtain the first polynomial; and obtaining a real-time control parameter gain according to the first polynomial and the real-time longitudinal speed of the vehicle.

[0018] To obtain the control parameter gain in real time, a polynomial approximation calculation method is applied based on the control parameter gain obtained offline, and the calculation value is optimized.

[0019] In any of the above embodiments, the step of obtaining the real-time control parameter gain according to the control parameter gain includes: taking the speed as an independent variable and performing a polynomial approximation of the control parameter gain of the third order, and the first polynomial is obtained in form:

[0020] Where a ₃₁ , a ₂₁ , a ₁₁ and a ₀₁ are the polynomial coefficients of the third order control parameters k ₁ , and ₃₂ , a ₂₂ , a ₁₂ and a ₀₂ are the polynomial coefficients of the third order control parameters k ₂ , and ₃₃ , a ₂₃ , a ₁₃ and a ₀₃ - polynomial coefficients of the parameters k ₃ control of the third order, and ₃₄ , a ₂₄ , a ₁₄ and a ₀₄ - polynomial coefficients of the parameters k ₄ control of the third order, v is the speed; obtaining the real-time control parameter gain K according to the first polynomial and the longitudinal speed of the vehicle in real time:

[0021] Where V _x is the longitudinal speed of the vehicle.

[0022] In this embodiment, a third-order polynomial approximation is applied, and the optimal control parameter gain is calculated in real time according to the actual speed fed back by the vehicle, which not only reduces the amount of calculation of the optimal control parameter gain, but also provides convergence and reliability of the optimal control parameter gain.

[0023] In any of the above embodiments, the step of obtaining a status error feedback value includes:

[0024] Obtaining a status error feedback value according to the following formula:

- скорость изменения отклонения бокового положения, e_ψ - отклонение курсового угла,

- скорость изменения отклонения курсового угла, X - величина обратной связи по ошибке состояния, V_x - продольная скорость транспортного средства, V_y - боковая скорость транспортного средства, (х,y) - положение транспортного средства в текущий момент, (x_des,y_des) - желаемое положение траектории, ψ - курсовой угол транспортного средства в текущий момент, ψ_des - курсовой угол желаемой траектории, ω - скорость рыскания транспортного средства и ρ - кривизна желаемой целевой точки.[0025] Where e _y is the deviation of the lateral position,

- rate of change of lateral position deviation, e _ψ - heading angle deviation,

- the rate of change of the heading angle deviation, X - the value of the status error feedback, V _x - the longitudinal speed of the vehicle, V _y - the lateral speed of the vehicle, (x,y) - the position of the vehicle at the current moment, (x _des ,y _des ) is the desired position of the trajectory, ψ is the heading angle of the vehicle at the current moment, ψ _des is the heading angle of the desired trajectory, ω is the yaw rate of the vehicle, and ρ is the curvature of the desired target point.

[0026] By updating the state error and passing it back to the control system gain, the side control effect becomes more accurate.

[0027] In any of the above embodiments, the step of obtaining a control quantity and a compensation quantity according to a control parameter gain, a real-time control parameter gain, a state error, a path curvature, and a vehicle pitch angle includes:

[0028] Obtaining a feedback control value in accordance with the real-time control parameter gain and the state error feedback value of:

δ _{sw_b} =-KX

[0029] where K is the real-time control parameter gain, X is the state error feedback amount;

[0030] obtaining a control value of the total position error in accordance with the total position error of:

- величина периода управления предыдущего цикла, составляющее δ_{sw_i}.[0031] where k _i is the summation factor, e _y is the lateral position deviation, T is the control period,

- the value of the control period of the previous cycle, which is δ _{sw_i} .

[0032] Obtaining the road curvature forward compensation value according to the path curvature of:

[0033] where m is the mass of the vehicle, V _x is the longitudinal speed of the vehicle, I _f is the distance from the front axle of the vehicle to the center of mass, I _r is the distance from the rear axle of the vehicle to the center of mass, C _f is the stiffness of the front wheel when cornering, C _r - rigidity of the rear wheel when cornering, R _des - radius of the trajectory, L - the sum of I _f and I _r ;

[0034] Obtaining the value of forward compensation of the slope of the road in accordance with the angle of the vehicle, component:

[0035] where A(i,j) is the parameter corresponding to the i-th row and j-th column of matrix A, B(i,j) is the parameter corresponding to the i-th row and j-th column of matrix B, g is the acceleration gravity, γ - the angle of inclination of the vehicle.

[0036] Specific methods for obtaining a feedback control quantity, a position sum error control quantity, a road curvature forward compensation quantity, a road slope forward compensation quantity are represented by the above formulas. In this embodiment, the road curvature determined by the desired trajectory is used to perform forward road curvature compensation, position information is used to perform forward road slope compensation, and a position sum deviation element is introduced to reduce the position error. And comprehensively take into account the influence of the uncertainty of the curvature of the road, the slope of the road and the general parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0037] In any of the above embodiments, the step of obtaining the expected steering angle according to the control variable and the offset value, and outputting the expected steering angle to the steering-by-wire system includes:

[0038] obtaining the expected steering angle according to the following formula:

[0039] where δ _{sw_b} is the feedback control value, δ _{sw_i} is the total position error control value, δ _{sw_c} is the road curvature forward compensation value, δ _{sw_r} is the road slope forward compensation value, I is the steering wheel ratio.

[0040] According to the calculation results of the feedback control quantity, the road curvature forward compensation quantity, the road slope forward compensation quantity, and the total position control quantity, the calculation results are summed and then multiplied by the steering gear ratio to obtain the final expected steering angle, which is output to the steering-by-wire system to provide trajectory tracking to ensure that the steady state error can still be completely eliminated if the overall vehicle parameters are not determined.

[0041] To solve the second object of the present invention, embodiments of the present invention provide a device for controlling the lateral movement of a self-driving vehicle, comprising: a setting module; the first acquisition module; a second acquisition module; a third acquisition module; a fourth acquisition module; a fifth acquisition module and a sixth acquisition module, wherein the linear quadratic controller control parameters are set by the setting module, the first acquisition module is configured to obtain general vehicle parameters, the second acquisition module is configured to obtain a control parameter gain in accordance with the control parameters and the general parameters vehicle, the second acquisition module is configured to obtain the control parameter gain in accordance with the control parameters and the general parameters of the vehicle, the third acquisition module is configured to obtain the control parameter gain in real time in accordance with the control parameter gain, the fourth module acquisition module is configured to obtain state error, trajectory curvature and vehicle inclination angle, the fifth acquisition module is configured to obtain the control variable and the compensation variable according to the control parameter gain, the real-time control parameter gain, the state error, the path curvature and the vehicle inclination angle, the sixth acquisition module is configured to obtain the expected steering angle according to the control variable and a compensating value, and outputting the expected steering angle to the steering-by-wire system.

[0042] This embodiment makes it possible to realize the optimal control of the lateral motion of a self-driving vehicle without static error, significantly reduce the amount of calculations, ensure the convergence of the optimal gain of the control system, and take into account the influence of road curvature, road slope and uncertainty of the overall parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0043] To solve the third object of the present invention, an embodiment of the present invention provides a self-driving vehicle, comprising: a vehicle body and a control device; moreover, the control device is configured to control the movement of the vehicle body using the method of controlling the lateral movement of a self-driving vehicle, as described in any of the embodiments of the present invention.

[0044] The self-driving vehicle provided by the embodiments of the present invention implements the steps of the self-driving vehicle lateral movement control method as in any embodiment of the present invention. Therefore, it has all the beneficial effects of the self-driving vehicle lateral control method, as in any embodiment of the present invention, which will not be repeated here.

[0045] Additional aspects and advantages of the present invention will become apparent in the following description or will be learned by putting the present invention into practice.

Brief description of the drawings

[0046] Additional aspects and advantages of the present invention will become apparent in the following description or will be learned by putting the present invention into practice.

[0047] FIG. 1 is a schematic representation of the lateral control of a self-driving vehicle of the related art;

[0048] in FIG. 2 is a flowchart of a self-driving vehicle lateral control method of one embodiment of the present invention;

[0049] in FIG. 3 is a flowchart of a method for obtaining a real-time control parameter gain of one embodiment of the present invention;

[0050] in FIG. 4 is a flowchart of a method for obtaining a control variable and a compensating variable of one embodiment of the present invention;

[0051] in FIG. 5 is a block functional diagram of a device for controlling the lateral movement of a self-driving vehicle of one embodiment of the present invention;

[0052] in FIG. 6 is a block functional diagram of a self-driving vehicle of one embodiment of the present invention;

[0053] in FIG. 7 is a schematic representation of the lateral control of a self-driving vehicle of a specific embodiment of the present invention; and

[0054] in FIG. 8 is a flowchart of a self-driving vehicle lateral control method of a particular embodiment of the present invention.

[0055] The corresponding relationship between reference signs and component names in FIG. 1-8 is as follows:

[0056] 100: Self-driving vehicle lateral movement control device, 110: setting module, 120: first acquisition module, 130: second acquisition module, 140: third acquisition module, 150: fourth acquisition module, 160: fifth acquisition module, 170 : sixth acquisition module, 200: self-driving vehicle, 210: vehicle body, 220: control device.

Implementation of the invention

[0057] In order that the above objects, features and advantages of the present invention can be understood more clearly, further detailed description of the present invention will be given below in connection with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other if there is no conflict.

[0058] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may also be practiced in other ways than those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0059] A method and apparatus for controlling the lateral movement of a self-driving vehicle in some embodiments of the present invention is described below with reference to FIG. 1-8.

[0060] The lateral motion control system of the self-driving vehicle calculates the desired steering command in real time according to the planned trajectory and the upper level information about the position of the vehicle, and controls the vehicle to control the movement along the target trajectory. Currently, linear quadratic optimal control (LQR) is widely used to control the lateral movement of a self-driving vehicle. As shown in FIG. 1, there are mainly the following problems:

[0061] (1) The optimal control gain is usually obtained by solving the Riccati equation online, which requires more computational resources and does not necessarily guarantee the convergence of the control gain.

[0062] (2) Currently, most self-driving vehicle lateral motion control systems use the implementation of forward road curvature compensation to reduce the steady state error, and do not take into account the lateral steady state error caused by the road slope, which reduces the quality of the vehicle lateral control system.

[0063] (3) There is a certain uncertainty in the general parameters of the vehicle, and the complete elimination of the error of the steady state lateral control of the vehicle cannot be guaranteed.

[0064] The related art is a parametrically adaptive LQR lateral motion control method for autonomous vehicles. The method of adjusting the LQR control parameters based on the path tracking error and the relationship between the position of the vehicle and the road determines the controller parameters in the current state. According to the determined control parameters, a steering quantity for steering the autonomous vehicle is calculated, and it is transmitted to the steering actuator for execution, which differs from this embodiment in that:

[0065] (1) Parameter adjustment methods vary. This related technology uses a path tracking error parameter and vehicle-road position relationship parameter adjustment method, and in this embodiment, a speed parameter adjustment method is used.

[0066] (2) The process for calculating the optimal gain of the control system is different. This related technology solves the Riccati equation online to obtain the optimum control system gain. In this embodiment, the optimal gain of the control system is calculated offline, and then a polynomial approximation is used to control the gain and update it in real time.

[0067] (3) The steering angle is different. This corresponding technique only considers the state error feedback control value. Based on the state error feedback control quantity, this embodiment also considers the road curvature forward compensation value, the road slope forward compensation value, and the total position error control quantity.

[0068] In general, the object of this embodiment is to solve at least one of the following problems.

[0069] (1) Solving the Riccati equation online to obtain the optimal control system gain is computationally intensive.

[0070] (2) The side steady state error caused by the slope of the road is not taken into account.

[0071] Embodiment 1

[0072] As shown in FIG. 2, this embodiment provides a method for controlling the lateral movement of a self-driving vehicle, including:

[0073] Step S102 setting the control parameters of the linear-quadratic controller;

[0074] Step S104 obtain general vehicle parameters;

[0075] Step S106 obtain control parameter gain in accordance with control parameters and vehicle general parameters;

[0076] Step S108 to obtain a gain of the control parameter in real time, in accordance with the gain of the control parameter;

[0077] Step S110 of obtaining the error feedback amount of the state, path curvature, and vehicle pitch angle;

[0078] Step S112, obtaining a control quantity and a compensation quantity according to the real-time control parameter gain, the state error feedback quantity, the path curvature, and the vehicle pitch angle; and

[0079] Step S114 of obtaining the expected steering angle according to the control quantity and the compensation value, and outputting the expected steering angle to the steering by wire system.

[0080] This embodiment makes it possible to realize the optimal control of the lateral movement of a self-driving vehicle without static error, significantly reduce the amount of calculations, ensure the convergence of the optimal gain of the control system, and take into account the influence of road curvature, road slope and uncertainty of the overall parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0081] Embodiment 2

[0082] In addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0083] In the above embodiment, the step of setting the control parameters of the linear-quadratic controller includes: the control parameters include the first weight matrix Q and the second weight matrix R, Q=diag[q ₁ , q ₂ , q ₃ , q ₄ ], R=[ r], where q ₁ , q ₂ , q ₃ , q ₄ and r are five control parameters, respectively, with q ₂ , q ₄ and r set to fixed values, aq ₁ , q ₃ are obtained in accordance with the following formula:

[0084] Where q _1max is the maximum value of q ₁ , q _1min is the minimum value of q ₁ , q _3max is the maximum value of q ₃ , q _3min is the minimum value of q ₃ , V _x is the longitudinal speed of the vehicle, V ₁ is the first threshold value speed, and V ₂ - the second threshold speed.

[0085] In this embodiment, q ₂ , q ₄ and r are set to fixed values, only for q ₁ , q ₃ , linear parameter adjustment is performed in accordance with the longitudinal speed V _x of the vehicle, that is, the weight matrix of control parameters is linear Quadratic controller uses a linear speed control method, which can reduce the amount of calculation.

[0086] Embodiment 3

[0087] In addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0088] The step of obtaining the control parameter gain according to the control parameters and the general vehicle parameters includes: obtaining the control parameter gain according to the following formula:

[0089] Where m is the mass of the vehicle, I _z is the inertia of rotation of the vehicle about the vertical direction, I _f is the distance from the front axle of the vehicle to the center of mass, I _r is the distance from the rear axle of the vehicle to the center of mass, C _f is front wheel cornering stiffness and C _r rear wheel cornering stiffness V _x vehicle longitudinal speed k ₁₄ control gain convergence matrix k ₁₄₍₁₎ , k ₁₄₍₂₎ , k _{14 (3)} and k ₁₄₍₄₎ - four elements in the convergence matrix of the gain of the control system, respectively, k ₁ , k ₂ , k ₃ and k ₄ - four gains of the control parameter, respectively, k ₁₄ - convergence matrix of the optimal gain control system computed using MATLAB, and k ₁ , k ₂ , k ₃ and k ₄ are four control parameter gains, respectively.

[0090] Moreover, the general parameters of the vehicle include the mass of the vehicle and / or the inertia of rotation of the vehicle around the vertical direction and / or the distance from the front axle of the vehicle to the center of mass and / or the distance from the rear axle of the vehicle to the center of mass and / or stiffness front wheels when cornering and/or rear wheel stiffness when cornering. The general vehicle parameters are obtained, the offline control parameter gain is obtained in accordance with the general vehicle parameters, and the calculation complexity of the method of this embodiment is reduced.

[0091] In accordance with the general parameters of the vehicle and the model of the controlled object, the optimal coefficients k ₁ , k ₂ , k ₃ and k ₄ control parameter gains are calculated offline using the Iqr function in MATLAB, which significantly reduces the amount of calculations compared to by solving the gain of the control parameters online.

[0092] Embodiment 4

[0093] As shown in FIG. 3, in addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0094] The step of obtaining the real-time control parameter gain according to the control parameter gain includes:

[0095] Step S202 of taking the speed as an independent variable and performing a polynomial approximation of the gain of the control parameter to obtain the first polynomial; and

[0096] Step S204 of obtaining the real-time control parameter gain according to the first polynomial and the real-time longitudinal speed of the vehicle.

[0097] To obtain the control parameter gain in real time, a polynomial approximation calculation method is applied based on the control parameter gain obtained offline, and the amount of calculation is optimized.

[0098] Embodiment 5

[0099] In addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0100] The step of obtaining the real-time control parameter gain in accordance with the control parameter gain includes: taking the speed as an independent variable and performing a polynomial approximation of the control parameter gain of the third order, and the first polynomial is obtained as:

[0101] Where a ₃₁ , a ₂₁ , a ₁₁ and a ₀₁ are the polynomial coefficients of the parameters k ₁ of the third order control, and ₃₂ , a ₂₂ , a ₁₂ and a ₀₂ are the polynomial coefficients of the parameters k ₂ of the third order control, and ₃₃ , and ₂₃ , a ₁₃ and a ₀₃ - polynomial coefficients of the parameters k ₃ control of the third order, and ₃₄ , a ₂₄ , a ₁₄ and a ₀₄ - polynomial coefficients of the parameters k ₄ control of the third order, v is the speed; obtaining the real-time control parameter gain K according to the first polynomial and the longitudinal speed of the vehicle in real time:

[0102] Where V _x is the longitudinal speed of the vehicle.

[0103] In this embodiment, a third-order polynomial approximation is applied, and the optimal control parameter gain is calculated in real time according to the actual speed transmitted back by the vehicle, which not only reduces the amount of calculation of the optimal control parameter gain, but also ensures convergence and reliability of the optimal control parameter gain.

[0104] Embodiment 6

[0105] In addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0106] The step of obtaining a status error feedback value includes:

[0107] Obtaining the status error feedback value according to the following formula:

- скорость изменения отклонения бокового положения, e_ψ - отклонение курсового угла,

- скорость изменения отклонения курсового угла, X - величина обратной связи по ошибке состояния, V_x - продольная скорость транспортного средства, V_y - боковая скорость транспортного средства, (x,y) - положение транспортного средства в текущий момент, (x_des,y_des) - желаемое положение траектории, ψ - курсовой угол транспортного средства в текущий момент, ψ_des - курсовой угол желаемой траектории, ω - скорость рыскания транспортного средства, и ρ - кривизна желаемой целевой точки.[0108] Where e _y is the deviation of the lateral position,

- rate of change of lateral position deviation, e _ψ - heading angle deviation,

- the rate of change of the course angle deviation, X - the value of the status error feedback, V _x - the longitudinal speed of the vehicle, V _y - the lateral speed of the vehicle, (x,y) - the position of the vehicle at the current moment, (x _des ,y _des ) is the desired position of the trajectory, ψ is the current heading of the vehicle, ψ _des is the heading of the desired trajectory, ω is the yaw rate of the vehicle, and ρ is the curvature of the desired target point.

[0109] By updating the state error and passing it back to the control system gain, the side control effect becomes more accurate.

[0110] Embodiment 7

[0111] As shown in FIG. 4, in addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0112] The step of obtaining the control variable and the compensation variable according to the control parameter gain, the real-time control parameter gain, the state error, the path curvature, and the vehicle pitch angle includes:

[0113] Step S302 of obtaining a feedback control amount according to the real-time control parameter gain and the status error feedback amount of:

δ _{sw_b} =-KX

[0114] where K is the real-time control parameter gain, X is the state error feedback amount;

[0115] Step S304 of obtaining the control amount of the total position error in accordance with the total position error of:

- величина периода управления предыдущего цикла of δ_{sw_i};[0116] where k _i is the summation factor, e _y is the deviation of the lateral position, T is the control period,

- the value of the control period of the previous cycle of δ _{sw_i} ;

[0117] Step S306 of obtaining the road curvature forward compensation amount according to the path curvature of:

[0118] where m is the mass of the vehicle, V _x is the longitudinal speed of the vehicle, I _f is the distance from the front axle of the vehicle to the center of mass, I _r is the distance from the rear axle of the vehicle to the center of mass, C _f is the stiffness of the front wheel when cornering, C _r - stiffness of the rear wheel when cornering, R _des - radius of the trajectory, L - the sum of I _f and I _r ;

[0119] Step S308 of obtaining the forward slope compensation amount according to the vehicle inclination angle, of:

[0120] where A(i,j) is the parameter corresponding to the i-th row and j-th column of matrix A, B(i,j) is the parameter corresponding to the i-th row and j-th column of matrix B, g is the acceleration gravity, γ - the angle of inclination of the vehicle.

[0121] Specific methods for obtaining a feedback control quantity, a position sum error control quantity, a road curvature forward compensation quantity, a road slope forward compensation quantity are represented by the above formulas. In this embodiment, the curvature of the road determined by the desired trajectory is used to perform forward road curvature compensation, the position information is used to perform forward compensation of the road slope, and an element of the total position deviation is introduced to reduce the position error. And the influence of the uncertainty of the road curvature, the road slope and the general parameters of the vehicle is comprehensively taken into account to achieve high reliability and high accuracy in tracking the desired trajectory.

[0122] This embodiment introduces a position sum deviation element to reduce the position error to ensure that the steady state error can still be completely eliminated if the overall vehicle parameters are not determined.

[0123] Embodiment 8

[0124] In addition to the technical features of the above-described embodiments, this embodiment further includes the following technical features.

[0125] The step of obtaining the expected steering angle according to the control variable and the compensation variable, and outputting the expected steering angle to the steering-by-wire system includes:

[0126] Obtaining the expected steering angle according to the following formula:

[0127] Where δ _{sw_b} is the feedback control value, δ _{sw_i} is the total position error control value, δ _{sw_c} is the road curvature forward compensation value, δ _{sw_r} is the road slope forward compensation value, I is the steering wheel ratio.

[0128] According to the calculation results of the feedback control quantity, the road curvature forward compensation quantity, the road slope forward compensation quantity, and the total position control quantity, the calculation results are summed and then multiplied by the steering gear ratio to obtain the final expected steering angle, which is output to the steering-by-wire system to provide trajectory tracking to ensure that the steady state error can be completely eliminated when the general parameters of the vehicle are not determined.

[0129] Embodiment 9

[0130] As shown in FIG. 5, this embodiment provides a device 100 for controlling the lateral movement of a self-driving vehicle, comprising: a setting module 110; the first acquisition module 120; the second acquisition module 130; the third receiving module 140; the fourth acquisition module 150; a fifth acquisition unit 160 and a sixth acquisition unit 170, wherein the linear quadratic controller control parameters are set by the setting unit 110, the first acquisition unit 120 obtains the general vehicle parameters, the second acquisition unit 130 obtains the control parameter gain according to the control parameters and the general vehicle parameters means, the second obtaining module 130 obtains the gain of the control parameter in accordance with the control parameters and the general parameters of the vehicle, the third module 140 obtains the gain of the control parameter in real time in accordance with the gain of the control parameter, the fourth module 150 receives a status error , the curvature of the path and the angle of the vehicle, the fifth acquisition unit 160 obtains the control quantity and the compensation quantity in accordance with the gain of the control parameter, coefficient gain value of the control parameter in real time, the state error, the curvature of the trajectory, and the angle of the vehicle, the sixth acquisition unit 170 obtains the expected steering angle in accordance with the control quantity and the compensation quantity, and outputs the expected steering angle to the steering system according to wires.

[0131] This embodiment makes it possible to realize the optimal control of the lateral movement of a self-driving vehicle without static error, significantly reduce the amount of calculations, ensure the convergence of the optimal gain of the control system, and take into account the influence of road curvature, road slope and uncertainty of the overall parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0132] Embodiment 10

[0133] As shown in FIG. 6, this embodiment provides a self-driving vehicle 200, comprising: a vehicle body 210 and a control device 220; wherein the control device controls movement of the vehicle body 210 using a self-driving vehicle lateral movement control method as described in any of the embodiments of the present invention.

[0134] Specific embodiment

[0135] This embodiment provides a self-driving vehicle lateral motion control method (i.e., an optimal control method without lateral static error) that can sufficiently reduce the amount of calculation of the optimal control system gain and ensure that the control system gain converges, and also take into account the influence of road curvature uncertainty, road inclination and general parameters to achieve high reliability and high accuracy of tracking the desired trajectory.

[0136] As shown in FIG. 7, this embodiment first calculates the optimal control system gain offline, then uses a polynomial to select the control system gain and stores it in the calculation unit, and then updates the optimal control system gain and the real-time state error value according to at the speed transmitted back by the vehicle. The curvature of the road is preemptively compensated for by the curvature of the road determined by the desired trajectory, and the slope of the road is preemptively compensated for by the vehicle's tilt angle obtained from the positioning information. In addition, a position sum deviation element is introduced to reduce the position error to ensure that the steady state error can still be completely eliminated if the overall vehicle parameters are not determined. As shown in FIG. 8, the implementation steps are as follows:

[0137] Step S402 of calculating the optimal gain of the offline control system and the weight matrix, wherein the control parameters involved in the LQR lateral motion controller comprise a first weight matrix Q and a second weight matrix R. Specifically, there is Q=diag[ q ₁ , q ₂ , q ₃ , q ₄ ], R=[r], that is, a total of 5 parameters q ₁ , q ₂ , q ₃ , q ₄ and r; according to the physical quantity associated with the control parameters, their importance is determined, and q ₂ , q ₄ and r are set as fixed values, only for q ₁ , q ₃ the parameters are adjusted according to the speed of the vehicle.

[0138] Offline calculation: according to the general parameters of the vehicle and the model of the controlled object, the optimal offline control system gains k ₁ , k ₂ , k ₃ and k ₄ are calculated using the Iqr function in MATLAB.

[0139] Polynomial approximation step S404: Performing a third-order polynomial approximation of the optimal speed control system gain according to the calculation results of the optimal control system gain.

[0140] Gain updating step S406: Calculating an optimal control system gain K in real time according to the actual speed fed back from the vehicle's wheel speedometer.

, отклонения курсового угла e_ψ, скорости изменения отклонения курсового угла

, величины X обратной связи по ошибке состояния в соответствии с положением транспортного средства, скоростью и ориентацией, выдаваемой системой позиционирования в режиме реального времени, и ожидаемой информацией о планируемой траектории.[0141] Status error update step S408: calculation of lateral position deviation e _y , rate of change of lateral position deviation

, heading angle deviation e _ψ , rate of change of heading angle deviation

, state error feedback values X in accordance with the position of the vehicle, the speed and orientation provided by the real-time positioning system, and the expected information about the planned trajectory.

[0142] Feedback control amount calculation step S410: multiplying the optimal gain of the control system and the state error feedback amount to calculate the state error feedback amount.

δ _{sw_b} =-KX

[0143] Step S412 calculating road curvature compensation amount ahead: calculating the road curvature compensation amount according to the path curvature provided by the planning system.

[0144] Step S414 of calculating the forward slope compensation amount: calculating the forward slope compensation amount in accordance with the vehicle inclination angle provided by the positioning system.

[0145] Step S416 of calculating the total position error control amount: inputting the total position error and calculating the summation amount due to the position error using incremental summation.

[0146] Steering angle calculation step S420: summing the calculation results of the road curvature feedforward compensation control quantity, the road slope feedforward compensation quantity, and the total position control quantity according to the feedback control quantity, and then multiply by the steering wheel ratio to obtain the final expected steering angle and output to a steering-by-wire system implementing trajectory tracking.

- величина периода управления предыдущего цикла 0f δ_{sw_i}, A(i, j), B(i, j) - параметр, соответствующий i-й строке и j-му столбцу матриц А и В, соответственно, и g - ускорение силы тяжести.[0147] In the above formula, i is the steering gear ratio, m represents the mass of the vehicle; Iz represents the rotational inertia of the vehicle about the vertical direction; ω represents the yaw rate of the vehicle; I _f and Ir represent the distance from the front and rear axles of the vehicle to the center of mass, respectively; C _f and C _r are the cornering stiffness of the front and rear wheels, respectively, γ is the angle of inclination provided by positioning, ψ, ψ _des are the heading angle of the vehicle at the current moment and the heading angle of the desired trajectory, respectively, (x,y), (x _des ,y _des ) - position of the vehicle at the current moment and position of the expected trajectory, k _i - summation factor, T - control period,

is the value of the control period of the previous cycle 0f δ _{sw_i} , A(i, j), B(i, j) is the parameter corresponding to the i-th row and j-th column of matrices A and B, respectively, and g is the acceleration of gravity.

[0148] In summary, the beneficial effects of the embodiments of the present invention are as follows:

[0149] 1. This embodiment can realize the optimal control of the lateral movement of a self-driving vehicle without static error, greatly reduce the amount of computation, ensure the convergence of the optimal control system gain, and take into account the influence of road curvature, road slope and uncertainty of the overall parameters of the vehicle to achieve high reliability and high accuracy of tracking the desired trajectory.

[0150] 2. This embodiment provides a static error-free method and introduces a position sum deviation element to reduce the position error to ensure that the steady state error can still be completely eliminated when the overall vehicle parameters are uncertain.

[0151] 3. The optimal control system gain is calculated in real time according to the actual speed fed back by the vehicle, which not only reduces the amount of calculation of the optimal control system gain, but also ensures the convergence and reliability of the optimal control system gain.

[0152] In this application, the terms "first", "second" and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance; the term "plurality" refers to two or more than two, unless otherwise expressly specified. The terms "installed", "connected", "connection", "fixed", etc. should be understood in a broad sense. For example, "connected" may be a fixed connection, a releasable connection, or a one-piece connection; "connection" can be connected directly or connected indirectly through an intermediate connection. For those skilled in the art, the specific meanings of the above terms in the present invention may be understood according to specific situations.

[0153] In the description of the present invention, the term "multiple" refers to two or more, unless clearly defined otherwise, and it should be understood that the orientation or position relationships indicated by the terms "top", "bottom", "left", "right" and the like, are orientation or position relationships based on what is shown in the drawings, are merely for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the said device or block must have a certain direction, and are designed and are operated in a particular orientation, and therefore cannot be considered as a limitation of the present invention.

[0154] In the description of the present invention, descriptions of the expressions "one embodiment", "some embodiments", and "a specific embodiment", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment(s) or example(s) are contained in at least one embodiment or example. In the specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined as appropriate in any one or more embodiments or examples.

[0155] The above descriptions are only a preferred embodiment of the present invention, which are not used to limit the present invention. For a person skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. in the spirit and principle of the present invention, everything should be included in the scope of protection of the present invention.

Claims (67)

1. A method for controlling the lateral movement of a self-driving vehicle, including: setting the control parameters of a linear-quadratic controller; getting the general parameters of the vehicle; obtaining a gain of the control parameter in accordance with the control parameters and the general parameters of the vehicle; obtaining a gain of the control parameter in real time in accordance with the gain of the control parameter; obtaining a feedback value on the status error, the curvature of the trajectory and the angle of the vehicle; obtaining a control quantity and a compensation quantity according to a real-time control parameter gain, a state error feedback quantity, a path curvature, and a vehicle tilt angle; and obtaining an expected steering angle according to the control variable and the compensation variable, and outputting the expected steering angle to the steering-by-wire system. 2. The method of controlling the lateral movement of a self-driving vehicle according to claim 1, in which the step of setting the control parameters of a linear-quadratic controller includes: control parameters include the first weight matrix Q and the second weight matrix R, Q=diag[q ₁ , q ₂ , q ₃ , q ₄ ], R=[r], where q ₁ , q ₂ , q ₃ , q ₄ and r are five control parameters, respectively, with q ₂ , q ₄ and r set to fixed values, aq ₁ , q ₃ are obtained in accordance with the following formula:

where q _1max is the maximum value of q ₁ , q _1min is the minimum value of q ₁ , q _3max is the maximum value of q ₃ , q _3min is the minimum value of q ₃ , V _x is the longitudinal speed of the vehicle, V ₁ is the first threshold speed and V ₂ - the second threshold speed. 3. The method for controlling the lateral movement of a self-driving vehicle according to claim 1, wherein the step of obtaining the gain of the control parameter in accordance with the control parameters and the general parameters of the vehicle includes: the control parameter gain is obtained according to the following formula:

where m is the mass of the vehicle, I _z is the inertia of rotation of the vehicle around the vertical direction, l _f is the distance from the front axle of the vehicle to the center of mass, l _r is the distance from the rear axle of the vehicle to the center of mass, C _f is the stiffness of the front wheel when cornering and C _r is the stiffness of the rear wheel when cornering, V _x the longitudinal speed of the vehicle, k ₁₄ is the convergence matrix of the gain of the control system, k ₁₄₍₁₎ , k ₁₄₍₂₎ , k ₁₄₍₃₎ and k ₁₄₍₄₎ - four elements in the convergence matrix of the gain of the control system, respectively, k ₁ , k ₂ , k ₃ and k ₄ - four gain coefficients of the control parameter, respectively. 4. The method for controlling the lateral movement of the self-driving vehicle according to claim 1, wherein the step of obtaining a control parameter gain in real time in accordance with the control parameter gain includes: taking the speed as an independent variable and performing a polynomial approximation of the gain of the control parameter to obtain the first polynomial; and obtaining a real-time control parameter gain according to the first polynomial and the real-time longitudinal speed of the vehicle. 5. The method for controlling the lateral movement of a self-driving vehicle according to claim 4, wherein the step of obtaining a gain of the control parameter in real time in accordance with the gain of the control parameter includes: taking the speed as an independent variable and performing a polynomial approximation of the third-order control gain, and the first polynomial is obtained as:

where a ₃₁ , a ₂₁ , a ₁₁ and a ₀₁ are the polynomial coefficients of the parameters k ₁ of the third order control, a ₃₂ , and ₂₂ , and ₁₂ and a ₀₂ are the polynomial coefficients of the parameters k ₂ of the third order control, and ₃₃ , a ₂₃ , a ₁₃ and a ₀₃ - polynomial coefficients of the parameters k ₃ control of the third order, and ₃₄ , a ₂₄ , and ₁₄ and a ₀₄ - polynomial coefficients of the parameters k ₄ control of the third order, v - speed; obtaining the real-time control parameter gain K according to the first polynomial and the longitudinal speed of the vehicle in real time:

where V _x is the longitudinal speed of the vehicle. 6. The method for controlling the lateral movement of a self-driving vehicle according to claim 1, wherein the step of obtaining a state error feedback value includes: getting the status error feedback value according to the following formula:

where e _y is the deviation of the lateral position,

- rate of change of lateral position deviation, e _ψ - heading angle deviation,

- the rate of change of the course angle deviation, X - the amount of feedback on the status error, V _x - the longitudinal speed of the vehicle, V _y - the lateral speed of the vehicle, (x,y) - the position of the vehicle at the current moment, (x _des ,y _des ) is the desired position of the trajectory, ψ is the current heading of the vehicle, ψ _des is the heading of the desired trajectory, ω represents the yaw rate of the vehicle, and ρ represents the curvature of the desired target point. 7. The method for controlling the lateral movement of the self-driving vehicle according to claim 1, wherein the step of obtaining the control variable and the compensating variable according to the control parameter gain, the real-time control parameter gain, the state error, the path curvature, and the vehicle inclination angle includes: obtaining a feedback control value in accordance with the real-time control parameter gain and the state error feedback value, which is: δ _{sw_b} = -KX, where K is the gain of the control parameter in real time, X is the amount of status error feedback; obtaining the control value of the total position error in accordance with the total position error, which is:

, where k _i is the summation factor, e _y is the deviation of the lateral position, T is the control period,

- the value of the control period of the previous cycle, component δ _{sw_i} ; obtaining the value of the forward compensation of the curvature of the road in accordance with the curvature of the trajectory, which is:

, where m is the mass of the vehicle, V _x is the longitudinal speed of the vehicle, l _f is the distance from the front axle of the vehicle to the center of mass, l _r is the distance from the rear axle of the vehicle to the center of mass, C _f is the stiffness of the front wheel when cornering , C _r - rigidity of the rear wheel when cornering, R _des - radius of the trajectory, L - the sum of l _f and l _r ; and obtaining the amount of forward compensation of the road slope in accordance with the angle of the vehicle, component:

, where A(i,j) is the parameter corresponding to the i-th row and j-th column of matrix A, B(i,j) is the parameter corresponding to the i-th row and j-th column of matrix B, g is the acceleration due to gravity , γ - the angle of inclination of the vehicle. 8. The method for controlling the lateral movement of a self-driving vehicle according to claim 1, wherein the step of obtaining an expected steering angle in accordance with the control variable and the compensating variable, and outputting the expected steering angle to the steering-by-wire system includes: obtaining the expected steering angle according to the following formula:

, where δ _{sw_b} is the feedback control value,

- control value of the total position error,

- amount of road curvature anticipatory compensation, δ _{sw_r} - anticipatory road slope compensation amount, I - steering wheel ratio. 9. Device (100) for controlling the lateral movement of a self-driving vehicle, containing: installation module (110); the first receiving module (120); the second module (130) receiving; the third module (140) receiving; the fourth module (150) receiving; fifth module (160) receiving and the sixth module (170) receiving, moreover, the control parameters of the linear-quadratic controller are set by the installation module (110), the first acquisition module (120) is configured to obtain general vehicle parameters; the second obtaining module (130) is configured to obtain the control parameter gain in accordance with the control parameters and the general parameters of the vehicle; the third obtaining module (140) is configured to obtain the gain of the control parameter in real time in accordance with the gain of the control parameter; the fourth obtaining module (150) is configured to obtain the state error, path curvature and vehicle pitch angle; the fifth acquisition unit (160) is configured to obtain a control quantity and a compensation quantity according to the control parameter gain, real-time control parameter gain, state error, path curvature, and vehicle tilt angle; and the sixth acquisition module (70) is configured to obtain the expected steering angle according to the control quantity and the compensation quantity, and output the expected steering angle to the steering-by-wire system. 10. Self-driving vehicle (200) containing: body (210) of the vehicle; control device (220); moreover, the control device (220) is configured to control the movement of the body (210) of the vehicle using the method of controlling the lateral movement of a self-driving vehicle according to any one of paragraphs. 1-8.

Images