CN109318905B - Intelligent automobile path tracking hybrid control method - Google Patents

Intelligent automobile path tracking hybrid control method Download PDF

Info

Publication number
CN109318905B
CN109318905B CN201810959618.9A CN201810959618A CN109318905B CN 109318905 B CN109318905 B CN 109318905B CN 201810959618 A CN201810959618 A CN 201810959618A CN 109318905 B CN109318905 B CN 109318905B
Authority
CN
China
Prior art keywords
vehicle
control
controller
speed
output
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810959618.9A
Other languages
Chinese (zh)
Other versions
CN109318905A (en
Inventor
蔡英凤
李健
孙晓强
王海
陈龙
梁军
袁朝春
江浩斌
何友国
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu University
Original Assignee
Jiangsu University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangsu University filed Critical Jiangsu University
Priority to CN201810959618.9A priority Critical patent/CN109318905B/en
Publication of CN109318905A publication Critical patent/CN109318905A/en
Application granted granted Critical
Publication of CN109318905B publication Critical patent/CN109318905B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0002—Automatic control, details of type of controller or control system architecture
    • B60W2050/0008—Feedback, closed loop systems or details of feedback error signal
    • B60W2050/0011—Proportional Integral Differential [PID] controller
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0019—Control system elements or transfer functions
    • B60W2050/0022—Gains, weighting coefficients or weighting functions

Abstract

The invention discloses an intelligent automobile path tracking hybrid control method, and belongs to the technical field of intelligent vehicle transverse control. The path tracking hybrid control method comprises the following steps: firstly, establishing a vehicle transverse control preview kinematics model at a low speed; step two, establishing a lateral control dynamic model of the vehicle at a high speed; and step three, designing a path tracking hybrid controller, which comprises a transverse controller, a monitor and a switching stable fuzzy controller, wherein the transverse controller is designed based on PID control and model predictive control, the monitor determines a tracking mode based on longitudinal vehicle speed, and the switching stable fuzzy controller is designed based on a fuzzy control theory. The path tracking hybrid control method provided by the invention effectively coordinates the problem of the requirement of the transverse control performance of the intelligent automobile under the high-speed and low-speed working conditions, and improves the feasibility, the accuracy and the stability of the path tracking of the intelligent automobile.

Description

Intelligent automobile path tracking hybrid control method
Technical Field
The invention belongs to the field of intelligent vehicle motion control, relates to an intelligent vehicle transverse control method, and particularly relates to an intelligent vehicle path tracking hybrid control method.
Background
The transverse motion control is one of key technologies for realizing the autonomous driving of the intelligent automobile, wherein the path tracking is to control the automobile to always drive along an expected path through autonomous steering, and meanwhile, the driving safety and riding comfort of the automobile are ensured, so that the transverse motion control is an ultimate target for unmanned driving. The intelligent automobile requires that the transverse motion control system has accurate, efficient and reliable control performance under the working condition of large-scale running, but the traditional single control algorithm cannot effectively coordinate the control requirements of the autonomous steering control system under different working conditions. Meanwhile, the intelligent automobile autonomous steering system has high real-time requirement on control, and the design of the traditional controller is difficult to ensure the steering performance under different working conditions, so that the design of the controller is simple and easy to realize.
From the aspects of accuracy, stability, easy implementation and the like of the intelligent automobile transverse motion control, different working conditions have different control targets and side points, so that the overall comprehensive performance is optimal. For example, when the vehicle is in a low-speed running condition, the kinematic characteristics of the vehicle are more prominent, and when the vehicle is in a high-speed running condition, the kinematic characteristics of the vehicle have a greater influence on the running state of the vehicle. The method has the advantages of simple design, good real-time performance and easy realization under the low-speed working condition, but cannot meet the reliability requirement of intelligent automobile path tracking under the high-speed working condition. The method comprises the following steps of researching the transverse control problem of the intelligent automobile by adopting a model prediction control method, wherein the algorithm firstly predicts the future output state of an object and then determines the control action at the current moment, namely firstly predicting and then controlling; the control algorithm can restrain the vehicle dynamics under the high-speed working condition, not only can quickly and accurately track a target path, but also ensures the safety and stability of vehicle running, but also has more complex controller design, large integral calculation amount and relatively larger realization difficulty.
Disclosure of Invention
In order to solve the technical problems, the invention provides an intelligent automobile path tracking hybrid control strategy, and designs an intelligent automobile path tracking hybrid switching control strategy consisting of a preview PID feedback control law and a model prediction control law. When the vehicle speed is low, a preview PID feedback control law which is simple in calculation and easy to realize is adopted under the condition that the vehicle is relatively safe, so that the rapidity and the feasibility of path tracking control are improved; and when the vehicle speed is higher, the model prediction control algorithm is adopted to improve the safety, stability and control accuracy of vehicle path tracking by considering that the vehicle has the characteristics of strong nonlinearity, time variation, instability and the like. In addition, a switching mechanism with stable supervision is designed while a high-speed and low-speed control mode is identified through the speed of the vehicle, and a proper control algorithm is introduced into each working condition, so that the local control performance of the system can be met, the aim of overall optimization can be fulfilled, and the feasibility, accuracy and stability of intelligent automobile path tracking are improved.
In order to achieve the above object, the technical scheme of the invention is as follows:
an intelligent automobile path tracking hybrid control method comprises the following steps:
s1, establishing a vehicle transverse control preview kinematics model at a low speed;
s2, establishing a lateral control dynamic model of the vehicle at high speed;
and S3, designing a path tracking hybrid controller, which comprises a transverse controller, a supervisor and a switching stability fuzzy controller, wherein the transverse controller comprises a PID controller and a model prediction controller.
Further, the specific step of S1 is:
s1.1, establishing a vehicle transverse control preview kinematic model as follows:
Figure BDA0001773538670000027
Figure BDA0001773538670000021
Figure BDA0001773538670000022
where v is the vehicle center of mass velocity, β is the vehicle center of mass yaw angle, ω is the vehicle yaw rate, Xc、YcRespectively the horizontal and vertical of the vehicle centroid position in the global geodetic coordinate systemThe coordinates of the position of the object to be imaged,
Figure BDA0001773538670000023
is the included angle between the longitudinal axis of the vehicle and the abscissa axis;
s1.2, calculating the transverse deviation and the course deviation at the pre-aiming point according to the geometric relation between the vehicle and the reference path, wherein the expression is as follows:
Figure BDA0001773538670000024
in the formula, xeIs the distance between the vehicle and the pre-aiming point in the vehicle coordinate system, yeIs the lateral deviation between the vehicle and the aiming point under the vehicle coordinate system,
Figure BDA0001773538670000025
is the course deviation, X, between the vehicle and the pre-aiming point under the vehicle coordinate systemfIs the abscissa, Y, of the pre-aiming point in the geodetic coordinate systemfIs the ordinate of the preview point in the geodetic coordinate system,
Figure BDA0001773538670000026
the heading angle of the pre-aiming point under a geodetic coordinate system;
s1.3, the rule for selecting the pre-aiming distance caused by the speed change is as follows:
xe＝xe0+kv
in the formula, xe0And k is a proportionality coefficient, and is an initial pre-aiming distance between the vehicle and a pre-aiming point under a vehicle coordinate system.
Further, the vehicle lateral control dynamics model in S2 is:
Figure BDA0001773538670000031
Figure BDA0001773538670000032
Figure BDA0001773538670000033
Figure BDA0001773538670000034
Figure BDA0001773538670000035
wherein m is the vehicle mass, IzIs the moment of inertia of the vehicle around the z-axis, a and b are the distances from the center of mass to the front and rear axes, deltafIs the turning angle of the front wheels of the vehicle,
Figure BDA0001773538670000036
as the yaw angle of the vehicle, Ccf、CcrCornering stiffness of front and rear tires of a vehicle, Clf、ClrLongitudinal stiffness, S, of the front and rear tires of a vehiclef、SrX and y are the abscissa and ordinate of the vehicle in the vehicle coordinate system, and X, Y are the abscissa and ordinate of the vehicle in the geodetic coordinate system, respectively, in terms of the slip ratio of the front and rear tires of the vehicle.
Further, the S3 specifically includes: the transverse controller consists of a PID controller established based on a preview kinematics model and a model prediction controller established based on a dynamics model, the monitor judges a high-speed mode and a low-speed mode by identifying the speed of the vehicle, and the switching stable fuzzy controller is designed based on a fuzzy control theory.
Further, the PID controller adopts an incremental PID control algorithm, and the PID algorithm formula is as follows:
Δu(k)＝Kp[e(k)-e(k-1)]+Kie(k)+Kd[e(k)-2e(k-1)+e(k-2)]
u(k)＝u(k-1)+Δu(k)
in the formula, assuming that the sampling period is T, the deviation is e (k) at the time k; kpIs proportional coefficient, integral coefficient
Figure BDA0001773538670000037
Coefficient of differentiation
Figure BDA0001773538670000038
TiFor integration time, TdFor differential time, u (k) is the output of the computer at the kth sampling instant, and the deviation e (k) is the deviation from the lateral deviation yeDeviation from heading
Figure BDA0001773538670000039
And (3) obtaining a comprehensive deviation e after non-dimensionalization, wherein the comprehensive deviation is used as an input quantity of the PID controller, and an output quantity u of the comprehensive deviation is a front wheel turning angle delta.
Further, the specific establishment process of the model predictive controller is as follows:
A. in model predictive controller, state quantity selection
Figure BDA00017735386700000310
The control quantity is selected from u ═ delta]Establishing a linear time-varying discrete model:
Figure BDA00017735386700000311
wherein ξ (k) is a discretized state quantity, y (k) is an output quantity, u (k) is a control quantity, a (k) and b (k) are discretized coefficient matrices, a (k) is I + ta (T), b (k) is tb (T), T is a sampling period, and I is a unit matrix;
B. the prediction equation for deriving model predictive control is as follows:
the prediction equation is an important part in model prediction control, the output of the system in a future period of time needs to be calculated, and the formula is converted into:
Figure BDA0001773538670000041
where x (kt) is the transformed matrix.
Obtaining a new state space expression:
Figure BDA0001773538670000042
wherein each matrix is defined as follows:
Figure BDA0001773538670000043
all three are the coefficient matrix in the prediction time domain, and η (kt) is the system output in the prediction time domain.
If the predicted time domain of the system is NpControl time domain as NcIn which N isc≤NpDefining the system output at the moment k as:
Figure BDA0001773538670000044
defining the system input at time k as:
Figure BDA0001773538670000045
the output Y (k +1k) of the system at the future k time is expressed in the form of a matrix:
Y(k+1|k)＝ψkξ(k)+ΘkΔU(k)
in the formula psikAnd ΘkAll coefficient matrixes in the prediction time domain, and delta U (k) is a control increment matrix, and the expression is as follows:
Figure BDA0001773538670000046
Figure BDA0001773538670000051
C. constructing constraint conditions, and adding dynamic constraints of vehicles such as centroid slip angle constraint, tire slip angle constraint, road adhesion conditions and the like;
D. the design objective function is:
Figure BDA0001773538670000052
wherein η (t + i | t) - ηr(t + i | t) is the difference between the actual output and the reference path, and ρ is the weightThe weight coefficient, epsilon is a relaxation factor, Q and R are weight matrixes, and delta u is a control increment;
E. and (3) optimizing and solving, namely solving the constrained optimization problem by the controller in each control period:
Figure BDA0001773538670000053
after solving the above formula in each control period, obtaining a control sequence of a first front wheel corner of the system, then acting a first element of the control sequence on the actual system until the next sampling moment, and solving a new control sequence again according to a new system measurement value at the next sampling moment.
Further, the working process of the monitor is as follows: when the speed of the vehicle is less than 50km/h, the monitor identifies that the vehicle runs under a low-speed working condition, and at the moment, a low-speed working mode is adopted, and when the speed of the vehicle is more than or equal to 50km/h, the monitor identifies that the vehicle runs under a high-speed working condition, and at the moment, a high-speed working mode is adopted.
Further, the switching stability fuzzy controller is specifically: front wheel steering angle value delta output by switching stable fuzzy controller to model predictive controller and PID controller1，δ2Performing weighting process to forcibly limit the output amplitude thereof, wherein1，λ2Output weighting coefficients of the transverse controller and the final output delta of the hybrid controller respectively for switching stable fuzzy controller outputfAccording to the formula deltaf＝λ1δ1+λ2δ2And obtaining the weighting coefficient is mainly reflected in the design of the fuzzy rule.
The invention has the beneficial effects that:
the invention provides an intelligent automobile path tracking hybrid switching control strategy consisting of a preview PID feedback control law, a model prediction control law and a switching stability fuzzy controller with supervision, which effectively coordinates the problem of the requirement of the transverse control performance of an intelligent automobile under high and low speed working conditions.
Drawings
FIG. 1 is a block diagram of a hybrid control system of the present invention.
Fig. 2 is a coordinate conversion chart.
FIG. 3 is a monorail model of a vehicle.
Fig. 4 is a switching stability fuzzy controller.
Fig. 5 shows membership functions corresponding to the weighting coefficients.
FIG. 6 is a graph of simulated longitudinal velocity variation of the present invention.
Fig. 7 is a simulation result diagram of the present invention, fig. 7(a) is a comparison result diagram of the lane change path tracking effect, fig. 7(b) is a comparison result diagram of the lateral acceleration, and fig. 7(c) is a comparison result diagram of the yaw rate.
Fig. 8 is a graph showing results of an actual vehicle test according to the present invention, fig. 8(a) is a graph showing results of comparison of the lane change path tracking effect, fig. 8(b) is a graph showing results of comparison of lateral acceleration, and fig. 8(c) is a graph showing results of comparison of yaw rate.
Detailed Description
The implementation process of the invention is described in detail in the following with the technical scheme and the attached drawings:
the invention combines the system characteristic difference of the intelligent automobile under the low-speed and high-speed steering working condition, firstly respectively establishes a vehicle transverse control preview kinematics model under the low speed and a vehicle transverse control dynamics model under the high speed, then designs a control strategy based on the established dynamics model, adopts PID control under the low speed mode, adopts model predictive control under the high speed mode, a monitor determines a path tracking control mode through the vehicle speed, and further designs a switching stable fuzzy controller with stable monitoring, realizes the smooth switching of a transverse control system, and finally realizes the intelligent automobile path tracking hybrid control, and the hybrid control system block diagram is shown in figure 1, and the block diagram comprises the transverse controller, the monitor and the switching stable fuzzy controller.
S1, establishing a vehicle transverse control preview kinematic model at low speed
S1.1, establishing a vehicle kinematic model
Under the low-speed driving working condition of a good road surface, the dynamics problems of vehicle stability control and the like generally do not need to be considered, and a path tracking controller designed based on a kinematic model has reliable control performance, so that the transverse control preview kinematic model of the vehicle at the low speed is established as follows:
kinematic model of vehicle As shown in FIG. 2, the position coordinates of the center of mass of the vehicle in the global geodetic coordinate system are (X)c，Yc) The included angle between the longitudinal axis of the vehicle and the axis of abscissa is
Figure BDA0001773538670000061
The following vehicle kinematics equations are established using geometric principles:
Figure BDA0001773538670000071
Figure BDA0001773538670000072
Figure BDA0001773538670000073
where v is the vehicle center of mass velocity, β is the vehicle center of mass yaw angle, and ω is the vehicle yaw rate.
S1.2, calculating the transverse deviation and the course deviation at the pre-aiming point according to the geometric relation between the vehicle and the reference path
The relationship between the vehicle coordinate system OXY and the geodetic coordinate system OXY is shown in FIG. 2, and a certain point O on the road ahead of the vehicle is setfThe coordinate in the geodetic coordinate system is (X)f，Yf) The preview point OfThe included angle between the tangential direction of the tracking curve and the abscissa axis of the geodetic coordinate system is
Figure BDA0001773538670000074
The included angle with the abscissa axis of the vehicle coordinate system is
Figure BDA0001773538670000075
The aiming point O can be adjusted according to the geometrical relationship in FIG. 2fPosition in the geodetic coordinate system
Figure BDA0001773538670000076
Conversion to position in vehicle coordinate system
Figure BDA0001773538670000077
The conversion relationship is as follows:
Figure BDA0001773538670000078
in the formula, xeIs the distance between the vehicle and the pre-aiming point in the vehicle coordinate system, yeIs the lateral deviation between the vehicle and the aiming point under the vehicle coordinate system,
Figure BDA0001773538670000079
is the course deviation, X, between the vehicle and the pre-aiming point under the vehicle coordinate systemfIs the abscissa, Y, of the pre-aiming point in the geodetic coordinate systemfIs the ordinate of the preview point in the geodetic coordinate system,
Figure BDA00017735386700000710
the heading angle of the pre-aiming point in the geodetic coordinate system is shown.
S1.3, the rule for selecting the pre-aiming distance caused by the speed change is as follows:
considering that the speed of the vehicle is variable when the vehicle runs, the selection of the pre-aiming distance has a large influence on the pre-aiming tracking effect, and when the speed of the vehicle is low, the information of a road in front of the vehicle is not well utilized due to the large pre-aiming distance; when the vehicle speed is higher, a smaller pre-aiming distance can cause information loss of a part of future roads, so that the path tracking control effect is poor, and therefore the selection rule of the pre-aiming distance is as follows:
xe＝xe0+kv (3)
in the formula, xe0The initial pre-aiming distance between the vehicle and the pre-aiming point under the vehicle coordinate system,k is a scaling factor.
S2, establishing a vehicle dynamic model as follows:
the intelligent vehicle runs in a complex traffic environment at higher speed, and in order to improve the reliability of the intelligent vehicle running at high speed, a more accurate vehicle dynamic model is necessarily introduced into a controller, so that the vehicle lateral control dynamic model at high speed is established as follows:
in the process of path tracking, the intelligent vehicle is necessarily accompanied by the change of the longitudinal vehicle speed, the change of the transverse vehicle speed and the change of the yaw angular velocity of the vehicle, so that a simple and effective vehicle single-track model with longitudinal and transverse coupling is established when the whole vehicle dynamics modeling is carried out. The invention mainly aims to research that the expected path tracked by the vehicle has better tracking precision and driving stability, and the influence of the vehicle suspension characteristics on a vehicle system is ignored; and simplifies the dynamic constraints of the vehicle with the aim of reducing the amount of calculations. Therefore, when the dynamic modeling is carried out, the following assumptions are firstly provided:
(1) assuming that the vehicle is always running on a flat road surface;
(2) the vehicle and suspension system are rigid, ignoring vertical motion of the vehicle;
(3) describing vehicle motion with a single-track model, without regard to left and right load shifting;
(4) assuming that the tire works in a linear region, neglecting the longitudinal-transverse coupling relation of tire force;
(5) ignoring longitudinal and lateral aerodynamics;
(6) neglecting the steering system, the input to the steering control is the front wheel steering angle deltaf。
In conclusion, the invention finally builds a monorail vehicle model with three degrees of freedom including longitudinal movement, transverse movement and transverse swinging, and the schematic diagram is shown in fig. 3:
according to Newton's second law, stress balance equations of the vehicle mass center along the x axis, the y axis and around the z axis are respectively obtained as follows:
Figure BDA0001773538670000081
wherein m is the vehicle mass, IzIs the moment of inertia of the vehicle about the z-axis, a, b are the distances from the center of mass to the front and rear axes, respectively, FcfAnd FcrLateral forces, F, respectively to the front and rear tyres of the vehiclelfAnd FlrLongitudinal forces, delta, respectively, to the front and rear tyres of the vehiclefIs the turning angle of the front wheels of the vehicle,
Figure BDA0001773538670000082
is the vehicle yaw angle; f in FIG. 3xfAnd FxrForces, F, respectively, exerted on the front and rear tires of the vehicle in the x-directionyfAnd FyrThe force applied to the front and rear tires of the vehicle in the y direction.
According to the assumptions, the vehicle tyre operates in a linear region, in which the slip angle and the longitudinal slip ratio are small and the lateral acceleration a is smally≦ 0.4g, the longitudinal and lateral forces of the tire may be expressed as:
Figure BDA0001773538670000083
in the formula, Ccf，CcrCornering stiffness for front and rear tires of a vehicle; clf，ClrLongitudinal stiffness of front and rear tires of a vehicle; sf，SrThe slip ratio of the front and rear tires of the vehicle.
Because a vehicle dynamics model established by the formula (4) has more trigonometric functions, the model is more difficult to simplify, and the following approximate relationship can be adopted if the vehicle front wheel rotation angle and the tire slip angle are smaller:
cosθ≈1,sinθ≈θ,tanθ≈θ (6)
and finally, taking the conversion relation between the vehicle body coordinate system and the geodetic coordinate system into consideration, and taking the simplified result into the formula (4) to obtain a vehicle nonlinear dynamic model:
Figure BDA0001773538670000091
Figure BDA0001773538670000092
Figure BDA0001773538670000093
Figure BDA0001773538670000094
Figure BDA0001773538670000095
wherein m is the vehicle mass, IzIs the moment of inertia of the vehicle around the z-axis, a and b are the distances from the center of mass to the front and rear axes, deltafIs the turning angle of the front wheels of the vehicle,
Figure BDA0001773538670000096
as the yaw angle of the vehicle, Ccf、CcrCornering stiffness of front and rear tires of a vehicle, Clf、ClrLongitudinal stiffness, S, of the front and rear tires of a vehiclef、SrX and y are the abscissa and ordinate of the vehicle in the vehicle coordinate system, and X, Y are the abscissa and ordinate of the vehicle in the geodetic coordinate system, respectively, in terms of the slip ratio of the front and rear tires of the vehicle.
S3, designing a transverse controller
S3.1, designing a PID controller established based on a preview kinematics model
Designing a PID controller based on the previously established vehicle transverse control preview kinematic model, wherein the parameters in the PID controller have a proportionality coefficient KPIntegral coefficient KIDifferential coefficient KDIn the real vehicle test, K is foundPAnd KDTwo parameters have a great influence on the path tracking of the intelligent vehicle: larger proportionality coefficient KPThe following capability of the intelligent automobile on a curve path can be improved, but the linear road is easy to vibrate; large differential coefficient KDThe intelligent automobile can enter a curve in advance and follow a good curved path to be straightThe line road behaves unstably and even easily deviates from the runway.
The conventional PID algorithm formula is as follows:
Figure BDA0001773538670000097
wherein u is a control amount, KpIs a proportionality coefficient, KiIs the integral coefficient, KdIs a differential coefficient, and e (t) is a deviation value.
Because the computer control system is a time discrete control system, the PID algorithm needs to be discretized: the invention adopts an incremental PID control algorithm, and the PID algorithm formula is as follows:
Δu(k)＝Kp[e(k)-e(k-1)]+Kie(k)+Kd[e(k)-2e(k-1)+e(k-2)](9)
u(k)＝u(k-1)+Δu(k) (10)
in the formula, assuming that the sampling period is T, the deviation is e (k) at the time k; kpIs proportional coefficient, integral coefficient
Figure BDA0001773538670000098
Coefficient of differentiation
Figure BDA0001773538670000099
TiFor integration time, TdFor differential time, u (k) is the output of the computer at the kth sampling instant, and the deviation e (k) is the deviation from the lateral deviation yeDeviation from heading
Figure BDA00017735386700000910
And (3) obtaining a comprehensive deviation e after non-dimensionalization, wherein the comprehensive deviation is used as an input quantity of the PID controller, and an output quantity u of the comprehensive deviation is a front wheel turning angle delta.
Dimensionless processing and fusing the deviations as follows:
Figure BDA0001773538670000101
Figure BDA0001773538670000102
Figure BDA0001773538670000103
in the formula (I), the compound is shown in the specification,
Figure BDA0001773538670000104
respectively the transverse deviation and the course deviation after the non-dimensionalization processing; y isemax、yeminMaximum and minimum values of the lateral deviation, respectively;
Figure BDA0001773538670000105
respectively the maximum value and the minimum value of course deviation; e is the composite error; n is a weight coefficient.
S3.2, designing a model predictive controller established based on a dynamic model
S3.2.1, a linear time-varying model is established as follows:
the method adopts a model predictive control algorithm at a higher vehicle speed, introduces a vehicle dynamic model into a controller, takes an accurate dynamic model as a predictive model, can improve the predictive capability of the controller on the future behavior of the vehicle, thereby improving the control precision of vehicle path tracking, but the problem of slow speed of solving the control quantity of the traditional model predictive controller at the higher vehicle speed is solved. Compared with a nonlinear model predictive control algorithm, the linear time-varying model predictive control algorithm with the linear time-varying model as the predictive model is adopted, the calculation is relatively simple, and the real-time performance is better, so that the rapidity and the real-time performance of the control can be improved.
In model predictive controller, state quantity selection
Figure BDA0001773538670000106
The control quantity is selected from u ═ delta]The nonlinear vehicle dynamics model established in S2 is linearized by a state trajectory-oriented linearization method, and a linear time-varying equation is obtained as follows:
Figure BDA0001773538670000107
wherein
Figure BDA0001773538670000108
C＝(0,0,0,0,1,0)TAll the three are coefficient matrixes; and y is an output quantity.
Discretizing the expression (14) by a first-order difference quotient method to obtain a discrete state control expression:
Figure BDA0001773538670000109
where ξ (k) is the state quantity after discretization, y (k) is the output quantity, a (k) and b (k) are the coefficient matrices after discretization, and a (k) is I + ta (T), b (k) is tb (T), T is the sampling period, and I is the unit matrix.
After introducing the incremental model, the state control expression is as follows:
Figure BDA0001773538670000111
where Δ ξ (k) is the increment of the state quantity, and Δ u (k) is the increment of the control quantity.
S3.2.2, the derived model predictive control prediction equation is as follows:
and (3) deducing a prediction equation of model prediction control on the basis of the linear state space model, and calculating the state quantity and the output quantity of the system in a prediction time domain through the prediction equation.
The prediction equation is an important part in model prediction control, and the output of the system in a future period of time needs to be calculated. Firstly, converting the formula (15) into:
Figure BDA0001773538670000112
in the formula, x (k | t) is a converted matrix.
A new state space expression can be obtained:
Figure BDA0001773538670000113
wherein each matrix is defined as follows:
Figure BDA0001773538670000114
the above three terms are all coefficient matrices in the prediction domain, and η (k | t) is the system output in the prediction domain.
If the predicted time domain of the system is NpControl time domain as NcIn which N isc≤NpDefining the system output at the moment k as:
Figure BDA0001773538670000115
defining the system input at time k as:
Figure BDA0001773538670000116
the output Y (k +1| k) of the system at the future time k is expressed in the form of a matrix:
Y(k+1|k)＝ψkξ(k)+ΘkΔU(k) (21)
in the formula psikAnd ΘkAll coefficient matrixes in the prediction time domain, and delta U (k) is a control increment matrix, and the expression is as follows:
Figure BDA0001773538670000121
Figure BDA0001773538670000122
s3.2.3, construction constraints are as follows
When the model predictive controller is designed, the constraints of the control quantity and the control increment are considered, the dynamic constraint condition of the vehicle at a higher speed is considered to be stricter than that of the vehicle at a lower speed, and the vehicle dynamic constraints including the dynamic constraints of the vehicle such as the centroid slip angle constraint, the tire slip angle constraint and the road adhesion condition are added, so that the safety, the stability and the comfort of the vehicle in running can be further ensured through the constraints.
a. Centroid slip angle constraint
The centroid slip angle has a large influence on the stability of the vehicle, and therefore it is necessary to increase the centroid slip angle constraint. According to research, the mass center slip angle limit of the vehicle for stable running can reach +/-12 degrees on a dry asphalt pavement with good adhesion conditions, and the limit value is approximately +/-2 degrees on an ice and snow pavement with poor adhesion conditions. Therefore, when the vehicle normally runs, the centroid slip angle needs to be restrained as follows:
-12 ° < β < 12 ° (good pavement) (22)
-2 ° < β < 2 ° (snow-covered road) (23)
b. Tire cornering angle restraint
If the tire slip angle is too large, the tire adhesion tends to reach the adhesion limit, so that the vehicle tends to slip and lose stability. From the cornering characteristics of the tire, it is known that when the tire cornering angle does not exceed 5 °, the cornering angle and the cornering force are approximately linearly related. According to the small angle constraint provided in the process of constructing the dynamic model, the constraint condition of the front wheel side deflection angle is set as follows:
-3°＜αf＜3° (24)
c. constraint of attachment condition
The dynamic performance of the automobile is also influenced by the road adhesion coefficient, and when the road adhesion condition is good, the factor has little influence on the running of the automobile; when conditions are severe, vehicle dynamics and passenger comfort are affected. When the vehicle is running on the road, the longitudinal acceleration a of the vehicleyLateral acceleration axThe road surface adhesion coefficient μ has the following relationship:
Figure BDA0001773538670000131
to this end, all constraints are incorporated into the solution process of quadratic programming.
S3.2.4, the design objective function is as follows:
because the vehicle dynamics model has high complexity and a plurality of dynamics constraints are added, the situation that the optimal solution cannot be calculated within a specified time is likely to occur in the actual execution process of the controller. Therefore, when the objective function is designed, the relaxation factor epsilon is added, and the expression of the objective function is obtained as follows:
Figure BDA0001773538670000132
wherein η (t + i | t) - ηr(t + i | t) is the difference between the actual output and the reference path. ρ is the weight coefficient, ε is the relaxation factor, Q and R are the weight matrix, and Δ u is the control increment. The expression first item reflects the following ability of the system to the expected track, and the second item reflects the requirement of the system for smooth change of the control quantity, and the expression generally functions to enable the system to track the expected track quickly and smoothly in a specified time.
S3.2.5, optimizing and solving:
according to the constraint conditions and the objective function established in the foregoing, the controller needs to solve the optimization problem with constraint in each control cycle:
Figure BDA0001773538670000133
after solving equation (27) in each control period, the first control sequence of the system is obtained, then the first element of the control sequence is applied to the actual system until the next sampling time, and a new control sequence is solved again according to the new system measurement value at the next sampling time.
S4, design supervisor:
the switching index is the longitudinal speed of the vehicle, the switching point of high speed and low speed is generally set to be 45-55km/h, so the switching speed is set to be 50km/h, when the speed is less than 50km/h, the monitor identifies that the vehicle runs under the low-speed working condition, a low-speed working mode is adopted, when the speed is more than or equal to 50km/h, the monitor identifies that the vehicle runs under the high-speed working condition, and a high-speed working mode is adopted.
S5, designing a switching stable fuzzy controller specifically comprises the following steps:
s5.1, switching the front wheel steering angle value delta output by the stable fuzzy controller to the model predictive controller and the PID controller1，δ2Performing weighting process to forcibly limit the output amplitude thereof, wherein1，λ2The lateral controllers output weighting coefficients respectively for switching the output of the stable fuzzy controller, and the hybrid controller finally outputs the front wheel corner deltafObtained according to formula (28). In the switching process, the two weighting coefficients work simultaneously, and after the switching is finished, one weighting coefficient is 1, and the other weighting coefficient is 0, so that system disturbance and transient response caused by large jump of controller output during control mode switching are prevented, and smooth switching and stable supervision of a transverse control system are realized.
δf＝λ1δ1+λ2δ2(28)
S5.2, switching the output of the stable fuzzy controller into a weighting coefficient output by the transverse controller, and switching the input of the stable fuzzy controller: the feedback value output by the carsim vehicle model in fig. 1 is subtracted from the expected output value corresponding to the target path, and the difference value between the current output and the target output and the change rate of the difference value are adopted, for the high-speed and low-speed switching control process, the current output represents the output of the previous controller, the target output represents the output of the controller after the switching process is finished, the controller output weighting coefficient 1 is the output weighting coefficient of the previous controller, the controller output weighting coefficient 2 is the output weighting coefficient of the controller to be operated, and the structure diagram of the switching stable fuzzy controller shown in fig. 4 is established.
S5.3, the basic discourse domain of the input quantity output deviation e of the controller is [ -40,40 [ -40 [ ]]The ambiguity domain is { -2, -1,0,1,2}, the corresponding ambiguity subset is { NB, NS, ZO, PS, PB }, and the output offset change rate deHas a basic discourse field of [ -28,28]The fuzzy domain is { -1,0,1}, the corresponding fuzzy subset is { N, ZO, P }, and the input quantity adopts a Gaussian membership function:
Figure BDA0001773538670000141
in the formula, σ represents the width of the membership function, and c represents the center of the membership function.
The output quantity of the controller is two, and is the output weighting coefficient of the horizontal controller, so the basic discourse domain of the two is [0,1], the fuzzy domain is {0,1,2,3}, the corresponding fuzzy subset is { ZO, PS, PM, PB }, wherein NB, NM, NS, ZO, PS, PM, PB, N, P are respectively called as negative big, negative middle, negative small, zero, positive small, positive middle, positive big, negative, positive; the output quantities all adopt the discrete triangular membership function of FIG. 5.
An expert experience method is adopted to define fuzzy control rules, and the control rules are shown in tables 1 and 2:
table 1 switching stability fuzzy controller output weighting coefficient 1 control rule table
Figure BDA0001773538670000142
Figure BDA0001773538670000151
Table 2 switching stability fuzzy controller output weighting coefficient 2 control rule table
Figure BDA0001773538670000152
The fuzzy control rule is described as a representative case:
1) when the output difference is positive and the change rate of the difference is positive, the current output has larger deviation with the target output and the difference has an increasing trend, in order to ensure the continuity of system switching, the output weighting coefficient 1 of the controller is required to be larger, and the output weighting coefficient 2 of the controller is required to be smaller;
2) when the output difference is negative and the change rate of the difference is positive, the current output has larger deviation with the target output and the difference has an increasing trend, and similarly, in order to ensure the continuity of system switching, the output weighting coefficient 1 of the controller is also larger and the output weighting coefficient 2 of the controller is smaller;
3) when the output difference is zero and the change rate of the difference is also zero, the current output is very close to the target output at the moment, the change of the difference is stable, the switching process is about to be completed, the controller output weighting coefficient 1 is zero at the moment, and the controller output weighting coefficient 2 is the maximum;
4) when the output difference is negative and the change rate of the difference is negative, the current output has a small deviation from the target output and the difference is continuously reduced, in order to ensure the continuity of system switching, the output weighting coefficient 1 of the controller is moderate, and the output weighting coefficient 2 of the controller is moderate.
The defuzzification algorithm adopts a common gravity center method, and the gravity center method is to take the gravity center of an area enclosed by a fuzzy membership function curve and a horizontal coordinate as a final output numerical value of the controller.
The path tracking curve designed by the invention is a lane changing path, and the expression is as follows:
Figure BDA0001773538670000153
in the formula, d is the transverse displacement of the vehicle after the lane change is finished, l is the longitudinal displacement of the vehicle after the lane change is finished, and d is 4m, and l is 100 m.
FIG. 6 is a diagram of the variation of the longitudinal speed of the vehicle according to the simulation of the present invention, and the relationship between the longitudinal speed of the vehicle and the variation of the abscissa X is shown in FIG. 6.
Fig. 7 is a diagram showing a simulation result of an intelligent vehicle path tracking hybrid control strategy according to the present invention, fig. 7(a) is a diagram showing a comparison result of a path change tracking effect, fig. 7(b) is a diagram showing a comparison result of a lateral acceleration, and fig. 7(c) is a diagram showing a comparison result of a yaw rate. Fig. 7(a) shows that the hybrid path tracking control designed by the invention has better path tracking effect than a single PID controller, wherein the maximum deviation is 0.043m, and it can be seen that the actual running path of the vehicle can well track the target path, and the vehicle is easy to generate a small deviation in a curve, but can be eliminated quickly. Fig. 7(b) -7 (c) show that during the lane change of the vehicle, the path tracking hybrid control without the stable switching supervisory controller is easy to generate sudden jitter, and the lateral acceleration and the yaw rate change are steep and unstable. The path tracking hybrid control with the stable supervision switching controller has smaller fluctuation than single PID control, and the lateral acceleration and the yaw velocity change are relatively stable and both are in a safe range, so that the hybrid control strategy designed by the invention can control the vehicle to be in a good stable state in the path tracking process. The lane change working condition simulation result shows that the hybrid control strategy can control the vehicle to accurately and stably track a target path under the conditions of high longitudinal speed and low longitudinal speed, and can realize smooth and stable switching between two control algorithms to achieve a good control effect.
Fig. 8 is a graph showing results of an actual vehicle test of an intelligent vehicle path tracking hybrid control strategy according to the present invention, fig. 8(a) is a graph showing results of comparison of the path change tracking effect, fig. 8(b) is a graph showing results of comparison of lateral acceleration, and fig. 8(c) is a graph showing results of comparison of yaw rate. The lane change condition real vehicle test result shows that the hybrid controller can control the vehicle path tracking deviation within the range of +/-0.15 m, the lateral acceleration within the range of +/-0.28 g/s and the yaw rate within the range of +/-2.5 degrees/s. The intelligent automobile path tracking hybrid control strategy designed by the invention can control the vehicle to track the target path quickly and stably, has better tracking precision, and obtains better control effect compared with single PID control.
The above-mentioned detailed description is only a preferred embodiment of the present invention, which is only used to illustrate the design ideas and features of the present invention, and not to limit the present invention, and any modifications, equivalent substitutions, improvements, etc. made under the technical ideas of the present invention should be included in the protection scope of the present invention.

Claims (7)

1. An intelligent automobile path tracking hybrid control method is characterized by comprising the following steps:
s1, establishing a vehicle transverse control preview kinematics model at a low speed;
s2, establishing a lateral control dynamic model of the vehicle at high speed;
the vehicle lateral control dynamic model is as follows:
Figure FDA0002311189970000011
Figure FDA0002311189970000012
Figure FDA0002311189970000013
Figure FDA0002311189970000014
Figure FDA0002311189970000015
wherein m is the vehicle mass, IzIs the moment of inertia of the vehicle around the z-axis, a and b are the distances from the center of mass to the front and rear axes, deltafIs the turning angle of the front wheels of the vehicle,
Figure FDA0002311189970000016
as the yaw angle of the vehicle, Ccf、CcrCornering stiffness of front and rear tires of a vehicle, Clf、ClrIs the front of a vehicleLongitudinal stiffness of the rear tire, Sf、SrThe slip rates of front and rear tires of the vehicle are respectively shown as x and y, and respectively represent the horizontal and vertical coordinates of the vehicle under a vehicle coordinate system, and X, Y respectively represent the horizontal and vertical coordinates of the vehicle under a geodetic coordinate system;
and S3, designing a path tracking hybrid controller, which comprises a transverse controller, a supervisor and a switching stability fuzzy controller, wherein the transverse controller comprises a PID controller and a model prediction controller.
2. The intelligent vehicle path tracking hybrid control method according to claim 1, wherein the specific steps of S1 are as follows:
s1.1, establishing a vehicle transverse control preview kinematic model as follows:
Figure FDA0002311189970000017
Figure FDA0002311189970000018
Figure FDA0002311189970000019
where v is the vehicle center of mass velocity, β is the vehicle center of mass yaw angle, ω is the vehicle yaw rate, Xc、YcRespectively the abscissa and ordinate of the position of the vehicle centroid in the global geodetic coordinate system,
Figure FDA00023111899700000110
is the included angle between the longitudinal axis of the vehicle and the abscissa axis;
s1.2, calculating the transverse deviation and the course deviation at the pre-aiming point according to the geometric relation between the vehicle and the reference path, wherein the expression is as follows:
Figure FDA0002311189970000021
in the formula, xeIs the distance between the vehicle and the pre-aiming point in the vehicle coordinate system, yeIs the lateral deviation between the vehicle and the aiming point under the vehicle coordinate system,
Figure FDA0002311189970000022
is the course deviation, X, between the vehicle and the pre-aiming point under the vehicle coordinate systemfIs the abscissa, Y, of the pre-aiming point in the geodetic coordinate systemfIs the ordinate of the preview point in the geodetic coordinate system,
Figure FDA0002311189970000023
the heading angle of the pre-aiming point under a geodetic coordinate system;
s1.3, the rule for selecting the pre-aiming distance caused by the speed change is as follows:
xe＝xe0+kv
in the formula, xe0And k is a proportionality coefficient, and is an initial pre-aiming distance between the vehicle and a pre-aiming point under a vehicle coordinate system.
3. The intelligent vehicle path tracking hybrid control method according to claim 1, wherein the S3 specifically is: the transverse controller consists of a PID controller established based on a preview kinematics model and a model prediction controller established based on a dynamics model, the monitor judges a high-speed mode and a low-speed mode by identifying the speed of the vehicle, and the switching stable fuzzy controller is designed based on a fuzzy control theory.
4. The intelligent automobile path tracking hybrid control method according to claim 3, characterized in that the PID controller adopts an incremental PID control algorithm, and the formula of the PID algorithm is as follows:
Δu(k)＝Kp[e(k)-e(k-1)]+Kie(k)+Kd[e(k)-2e(k-1)+e(k-2)]
u(k)＝u(k-1)+Δu(k)
in the formula, assuming that the sampling period is T, the deviation is e (k) at the time k; kpIs proportional coefficient, integral coefficient
Figure FDA0002311189970000024
Coefficient of differentiation
Figure FDA0002311189970000025
TiFor integration time, TdFor differential time, u (k) is the output of the computer at the kth sampling instant, and the deviation e (k) is the deviation from the lateral deviation yeDeviation from heading
Figure FDA0002311189970000026
And (3) obtaining a comprehensive deviation e after non-dimensionalization, wherein the comprehensive deviation is used as an input quantity of the PID controller, and an output quantity u of the comprehensive deviation is a front wheel turning angle delta.
5. The intelligent automobile path tracking hybrid control method according to claim 3, wherein the specific establishment process of the model predictive controller is as follows:
A. in model predictive controller, state quantity selection
Figure FDA0002311189970000027
The control quantity is selected from u ═ delta]Establishing a linear time-varying discrete model:
Figure FDA0002311189970000028
wherein ξ (k) is a discretized state quantity, y (k) is an output quantity, u (k) is a control quantity, a (k) and b (k) are discretized coefficient matrices, a (k) is I + ta (T), b (k) is tb (T), T is a sampling period, and I is a unit matrix;
Figure FDA0002311189970000031
as is the longitudinal speed of the vehicle,
Figure FDA0002311189970000032
as the lateral speed of the vehicle,
Figure FDA0002311189970000033
in order to provide a yaw angle of the vehicle,
Figure FDA0002311189970000034
the yaw angular velocity of the vehicle is obtained, Y is the ordinate of the vehicle under the geodetic coordinate system, and X is the abscissa of the vehicle under the geodetic coordinate system;
B. the prediction equation for deriving model predictive control is as follows:
the prediction equation is an important part in model prediction control, the output of the system in a future period of time needs to be calculated, and the formula is converted into:
Figure FDA0002311189970000035
wherein x (kt) is a matrix obtained by combining and converting the state quantity at the time k and the control quantity at the time (k-1);
obtaining a new state space expression:
Figure FDA0002311189970000036
wherein each matrix is defined as follows:
Figure FDA0002311189970000037
all three are coefficient matrices in the prediction time domain, η (k | t) is the system output in the prediction time domain;
if the predicted time domain of the system is NpControl time domain as NcIn which N isc≤NpDefining the system output at the moment k as:
Figure FDA0002311189970000038
defining the system input at time k as:
Figure FDA0002311189970000039
the output Y (k +1| k) of the system at the future time k is expressed in the form of a matrix:
Y(k+1|k)＝ψkξ(k)+ΘkΔU(k)
wherein Y (k +1| k) represents the output quantity of the system at the (k +1) th future time predicted at the time k;
in the formula psikAnd ΘkAll coefficient matrixes in the prediction time domain, and delta U (k) is a control increment matrix, and the expression is as follows:
Figure FDA0002311189970000041
Figure FDA0002311189970000042
C. constructing constraint conditions, and adding centroid slip angle constraint, tire slip angle constraint and road surface attachment condition constraint;
D. the design objective function is:
Figure FDA0002311189970000043
wherein η (t + i | t) - ηr(t + i | t) is the difference between the actual output and the reference path, ρ is the weight coefficient, ε is the relaxation factor, Q and R are the weight matrices, Δ u is the control increment;
E. and (3) optimizing and solving, namely solving the constrained optimization problem by the controller in each control period:
Figure FDA0002311189970000044
after solving the above formula in each control period, obtaining a control sequence of a first front wheel corner of the system, then acting a first element of the control sequence on the actual system until the next sampling moment, and solving a new control sequence again according to a new system measurement value at the next sampling moment.
6. The intelligent automobile path tracking hybrid control method according to claim 1, wherein the working process of the supervisor is as follows: when the speed of the vehicle is less than 50km/h, the monitor identifies that the vehicle runs under a low-speed working condition, and at the moment, a low-speed working mode is adopted, and when the speed of the vehicle is more than or equal to 50km/h, the monitor identifies that the vehicle runs under a high-speed working condition, and at the moment, a high-speed working mode is adopted.
7. The intelligent vehicle path tracking hybrid control method according to claim 1, wherein the switching stability fuzzy controller is specifically: front wheel steering angle value delta output by switching stable fuzzy controller to model predictive controller and PID controller1，δ2Performing weighting process to forcibly limit the output amplitude thereof, wherein1，λ2The weight coefficients are output by the transverse controller and the front wheel turning angle delta finally output by the hybrid controller respectivelyfAccording to the formula deltaf＝λ1δ1+λ2δ2And obtaining the weighting coefficient through the output of the switching stable fuzzy controller.
CN201810959618.9A 2018-08-22 2018-08-22 Intelligent automobile path tracking hybrid control method Active CN109318905B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810959618.9A CN109318905B (en) 2018-08-22 2018-08-22 Intelligent automobile path tracking hybrid control method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810959618.9A CN109318905B (en) 2018-08-22 2018-08-22 Intelligent automobile path tracking hybrid control method

Publications (2)

Publication Number Publication Date
CN109318905A CN109318905A (en) 2019-02-12
CN109318905B true CN109318905B (en) 2020-06-09

Family

ID=65263872

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810959618.9A Active CN109318905B (en) 2018-08-22 2018-08-22 Intelligent automobile path tracking hybrid control method

Country Status (1)

Country Link
CN (1) CN109318905B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102022116993A1 (en) 2022-07-07 2024-01-18 Valeo Schalter Und Sensoren Gmbh METHOD FOR DETERMINING A CONDITION OF A VEHICLE

Families Citing this family (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109808707B (en) * 2019-02-19 2020-12-08 武汉理工大学 Automobile steering control method and controller based on stochastic model prediction
CN109884900B (en) * 2019-04-03 2022-04-12 东南大学 Design method of harvester path tracking controller based on adaptive model predictive control
CN110186470A (en) * 2019-04-26 2019-08-30 纵目科技（上海）股份有限公司 The reference line for meeting dynamics of vehicle generates system, terminal and application method
CN110210058A (en) * 2019-04-26 2019-09-06 纵目科技（上海）股份有限公司 Meet reference line generation method, system, terminal and the medium of dynamics of vehicle
CN110539752B (en) * 2019-06-26 2020-12-18 江苏大学 Intelligent automobile multi-prediction-range model prediction trajectory tracking control method and system
CN110398969B (en) * 2019-08-01 2022-09-27 北京主线科技有限公司 Domain steering control method and device during self-adaptive prediction of automatic driving vehicle
CN110687797B (en) * 2019-11-11 2022-04-19 湖南大学 Self-adaptive MPC parking transverse control method based on position and posture
CN110989625B (en) * 2019-12-25 2020-11-27 湖南大学 Vehicle path tracking control method
CN111532283B (en) * 2020-05-15 2022-03-25 吉林大学 Model prediction control-based path tracking method for semi-trailer train
CN111812974B (en) * 2020-05-28 2021-07-13 北京理工大学 Comprehensive control method for bilateral motor-driven tracked vehicle
CN111806445A (en) * 2020-05-29 2020-10-23 北汽福田汽车股份有限公司 Vehicle transverse control method and device, medium, equipment and vehicle
CN111845738B (en) * 2020-06-22 2021-10-12 江苏大学 Vehicle path tracking control method based on double-model combination
CN111703417B (en) * 2020-06-24 2023-09-05 湖北汽车工业学院 High-low speed unified pre-aiming sliding film driving control method and control system
CN111665850B (en) * 2020-06-29 2022-02-01 合肥工业大学 Track tracking transverse control method and device for unmanned formula racing car
CN111735228B (en) * 2020-06-30 2021-10-15 中船重工湖北海洋核能有限公司 Variable-structure control system and control method of lithium bromide refrigerator for marine nuclear power ship
CN113954833B (en) * 2020-07-06 2023-05-30 湖南工业大学 Full-electric-drive distributed unmanned vehicle path tracking and stability coordination control method
CN111959500B (en) * 2020-08-07 2022-11-11 长春工业大学 Automobile path tracking performance improving method based on tire force distribution
CN114074674B (en) * 2020-08-14 2024-04-09 上海汽车工业(集团)总公司 Method and related device for acquiring history track curve of guided vehicle
CN111923908A (en) * 2020-08-18 2020-11-13 哈尔滨理工大学 Stability-fused intelligent automobile path tracking control method
CN112193318A (en) * 2020-10-15 2021-01-08 北京航天发射技术研究所 Vehicle path control method, device, equipment and computer readable storage medium
CN114442601A (en) * 2020-11-06 2022-05-06 郑州宇通客车股份有限公司 Unmanned vehicle tracking control method and device
CN112859838B (en) * 2020-12-31 2022-05-17 清华大学苏州汽车研究院（吴江） Automatic driving control method, device, equipment and medium
CN112829766B (en) * 2021-02-07 2022-05-17 西南大学 Adaptive path tracking method based on distributed driving electric vehicle
CN113002620B (en) * 2021-03-12 2023-04-25 重庆长安汽车股份有限公司 Automatic steering wheel angle deviation correction method and system and vehicle
CN113204236B (en) * 2021-04-14 2022-05-20 华中科技大学 Intelligent agent path tracking control method
CN113340619B (en) * 2021-05-31 2022-07-29 青岛森麒麟轮胎股份有限公司 Method and system for evaluating tire braking performance
CN113525366A (en) * 2021-07-28 2021-10-22 日照公路建设有限公司 Transverse control method for hydraulic transverse controller of steel-wheel road roller
CN113665587A (en) * 2021-08-24 2021-11-19 东风柳州汽车有限公司 Lateral control method, device, storage medium, and apparatus for autonomous vehicle
CN113900438B (en) * 2021-10-08 2023-09-22 清华大学 Unmanned vehicle path tracking control method, unmanned vehicle path tracking control device, computer equipment and storage medium
CN113985868B (en) * 2021-10-09 2023-08-08 北京科技大学 Layered path tracking control implementation method for wheeled mobile robot
CN114355897B (en) * 2021-12-15 2023-08-29 同济大学 Vehicle path tracking control method based on model and reinforcement learning hybrid switching
CN114326728A (en) * 2021-12-24 2022-04-12 江苏大学 Single AGV intelligent garage path tracking control system and method with high safety margin
CN115346366B (en) * 2022-07-22 2023-11-28 武汉理工大学 Intelligent network coupled vehicle team control method and system considering road adhesion coefficient
CN115236972A (en) * 2022-07-25 2022-10-25 中国安全生产科学研究院 Multi-robot formation control method based on double closed-loop self-adaptive PID

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014091922A1 (en) * 2012-12-13 2014-06-19 独立行政法人国立高等専門学校機構 Power assist system using model predictive control
CN107097785A (en) * 2017-05-25 2017-08-29 江苏大学 A kind of adaptive intelligent vehicle crosswise joint method of preview distance
CN107804315A (en) * 2017-11-07 2018-03-16 吉林大学 It is a kind of to consider to drive people's car collaboration rotating direction control method that power is distributed in real time
CN108099900A (en) * 2017-12-18 2018-06-01 长春工业大学 The laterally stable four-wheel steering control method of automobile is kept under a kind of limiting condition
CN108248605A (en) * 2018-01-23 2018-07-06 重庆邮电大学 The transverse and longitudinal control method for coordinating that a kind of intelligent vehicle track follows

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014091922A1 (en) * 2012-12-13 2014-06-19 独立行政法人国立高等専門学校機構 Power assist system using model predictive control
CN107097785A (en) * 2017-05-25 2017-08-29 江苏大学 A kind of adaptive intelligent vehicle crosswise joint method of preview distance
CN107804315A (en) * 2017-11-07 2018-03-16 吉林大学 It is a kind of to consider to drive people's car collaboration rotating direction control method that power is distributed in real time
CN108099900A (en) * 2017-12-18 2018-06-01 长春工业大学 The laterally stable four-wheel steering control method of automobile is kept under a kind of limiting condition
CN108248605A (en) * 2018-01-23 2018-07-06 重庆邮电大学 The transverse and longitudinal control method for coordinating that a kind of intelligent vehicle track follows

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于横向与纵向综合控制的智能车辆运动控制研究;王浩;《南京航空航天大学》;20170404;第1，32-36,47-48页 *
智能车辆路径跟踪横向控制方法的研究;赵熙俊等;《汽车工程》;20110906;第382-387页 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102022116993A1 (en) 2022-07-07 2024-01-18 Valeo Schalter Und Sensoren Gmbh METHOD FOR DETERMINING A CONDITION OF A VEHICLE

Also Published As

Publication number Publication date
CN109318905A (en) 2019-02-12

Similar Documents

Publication Publication Date Title
CN109318905B (en) Intelligent automobile path tracking hybrid control method
CN107161207B (en) Intelligent automobile track tracking control system and control method based on active safety
CN112092815B (en) Vehicle track changing tracking control method based on model prediction
CN107943071B (en) Formation maintaining control method and system for unmanned vehicle
CN109799814B (en) Design method of track unmanned vehicle track tracking controller based on kinematics model
CN111497826B (en) Coordinated control method and system for yaw stability of electric automobile
CN111258323A (en) Intelligent vehicle trajectory planning and tracking combined control method
CN111923908A (en) Stability-fused intelligent automobile path tracking control method
CN111750866B (en) Intelligent automobile transverse and longitudinal coupling path planning method based on regional virtual force field
CN109017759B (en) Desired path vehicle yaw control method
CN112947469A (en) Automobile track-changing track planning and dynamic track tracking control method
CN109017804B (en) Method for distributing driving torque for each hub motor of vehicle by torque distribution controller
CN109017446B (en) Expected path vehicle longitudinal speed tracking control method and device
CN112578672A (en) Unmanned vehicle trajectory control system based on chassis nonlinearity and trajectory control method thereof
CN109017447B (en) Method for outputting total driving torque of unmanned vehicle
Chang et al. An adaptive MPC trajectory tracking algorithm for autonomous vehicles
CN111845738A (en) Vehicle path tracking control method based on double-model combination
Chen et al. An overtaking obstacle algorithm for autonomous driving based on dynamic trajectory planning
CN114684199A (en) Vehicle dynamics series hybrid model driven by mechanism analysis and data, intelligent automobile trajectory tracking control method and controller
CN114179818A (en) Intelligent automobile transverse control method based on adaptive preview time and sliding mode control
Li et al. An improved model predictive control method for vehicle lateral control
CN113183953A (en) Vehicle post-collision active safety control method and system based on distributed driving chassis
CN113867330B (en) Control method for realizing vehicle drift under any path based on multi-degree-of-freedom prediction model
Zang et al. Research on Intelligent Vehicle Path Planning Based on Improved Artificial Potential Field Method
Liu et al. Research on stability of high-speed autonomous vehicles based on linear time-varying model predictive control

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant