CN115292671A - Driver horizontal-vertical coupling behavior model - Google Patents
Driver horizontal-vertical coupling behavior model Download PDFInfo
- Publication number
- CN115292671A CN115292671A CN202211054892.4A CN202211054892A CN115292671A CN 115292671 A CN115292671 A CN 115292671A CN 202211054892 A CN202211054892 A CN 202211054892A CN 115292671 A CN115292671 A CN 115292671A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- driver
- longitudinal
- speed
- behavior
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
- G06F17/12—Simultaneous equations, e.g. systems of linear equations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
A driver transverse and longitudinal coupling behavior model belongs to the technical field of automobile simulation tests. The invention aims to consider the transverse and longitudinal behavior coupling relation of a driver, namely a driver behavior model capable of controlling speed and direction simultaneously, and introduces road curvature in the process of driving preview decision, so that the driver can properly adjust the vehicle speed according to the road curvature to ensure the safety of a curve. The model establishing process of the invention is as follows: establishing a driver preview behavior module, establishing a driver decision behavior module, establishing a driver forecast behavior module and establishing a driver optimization behavior module. In the process of model building, the vehicle is regarded as a two-dimensional entity instead of simple particles, and the rectangle represents the safe driving outline of the vehicle, so that the method is more practical.
Description
Technical Field
The invention belongs to the technical field of automobile simulation test.
Background
In recent years, automobiles have become an important tool for people to go out, and along with the important traffic safety problem, tens of thousands of people are lost in traffic accidents every year. The safety and stability of a qualified automobile need to be tested in a large quantity, which requires the participation of a driver, and the real automobile test is time-consuming and labor-consuming, so that the virtual test has become a research hotspot problem of automobile test and evaluation. When the relevant performance of the automobile is virtually tested, the automobile is driven by a virtual driver, so that some performance problems of the automobile are reflected, and the driving behavior of a real driver needs to be simulated by the virtual driver at the moment through a driver model.
The conventional driver model usually considers the steering control behavior of the driver or the longitudinal speed control behavior of the driver, namely the following behavior of the driver, wherein the former is to keep the longitudinal speed of the automobile constant, and the latter is to keep the driving direction of the automobile constant. The two driver models have certain application limitation
Disclosure of Invention
The invention aims to consider the transverse and longitudinal behavior coupling relation of a driver, namely a driver behavior model capable of controlling speed and direction simultaneously, and introduces road curvature in the process of driving preview decision, so that the driver can properly adjust the vehicle speed according to the road curvature to ensure the safety of a curve.
The model establishing process of the invention is as follows:
s1, establishing a driver preview behavior module
After preliminarily discretizing the transverse and longitudinal accelerations, pre-aiming time T is carried out on the transverse and longitudinal accelerations of each group p The prediction of the motion track in the method is that the preview time is divided into m equal parts
And assuming that the transverse and longitudinal acceleration and the centroid slip angle of the vehicle in the prealignment time are fixed values, and utilizing the following formula to carry out vehicle state information in the prealignment time:
and updating the position information of the vehicle using the following formula:
wherein X and Y are the horizontal and vertical coordinates of the mass center of the vehicle under the geodetic coordinate system，v x And v y Is the transverse and longitudinal speed of the vehicle in the vehicle coordinate system, A x And A y Is the transverse and longitudinal acceleration of the vehicle,
the heading angle of the vehicle is shown, omega is the yaw angular velocity of the vehicle, beta is the mass center side slip angle of the vehicle, and j =1,2,3 \8230m;
the safety constraint is divided into two parts, namely road boundary safety constraint and curve safety vehicle speed constraint;
and calculating the coordinates of four vertexes of the safety rectangle according to the coordinates of the mass center of the vehicle and the course angle by using the following formula:
wherein X A And Y A Is the horizontal and vertical coordinates of the safe rectangle vertex A, and the like, X and Y are respectively the horizontal and vertical coordinates of the vehicle mass center,
l and W are the length and width of the vehicle safety rectangle; whether the vehicle collides with the road boundary can be converted into the problem of the position relation between the rectangle and the simple polygon through the four vertex coordinates and the centroid coordinates of the safety rectangle, if the vehicle does not collide with the road boundary, the vehicle is kept, otherwise, the motion trail is deleted;
and (3) safety vehicle speed restraint at the bend: at a curve where the vehicle is traveling, the following relationship exists:
wherein F is the lateral force of the vehicle at the curve, V is the running speed of the vehicle, R is the curvature radius of the road, mu is the friction coefficient of the road, and theta is the inclination angle of the road;
when the road inclination angle is zero, the speed safety constraint of the vehicle in the curve can be obtained by introducing the style coefficient epsilonThe following were used:
in the driving process of the vehicle, the maximum safe speed of the vehicle in the curve which is considered by a driver to drive can be obtained through calculating the real-time road curvature, and then the vehicle speed can be correspondingly adjusted to ensure that the vehicle safely passes through the curve;
s2, establishing a driver decision behavior module
Making a decision on a feasible region according to three aspects of driving stability, driving efficiency and operation smoothness, and respectively establishing the following objective functions:
wherein, J p An objective function representing the driving stationarity, d j The distance from the mass center of the vehicle to the center line of the road at the jth point in the motion trail prediction is obtained; j is a unit of e An objective function representing driving efficiency, A x Is the vehicle longitudinal acceleration; j. the design is a square cx And J cy Representing ride smoothness, Δ A x And Δ A y Representing the variation of the horizontal and vertical acceleration, namely the horizontal and vertical acceleration variation value of the decision period is larger than that of the previous decision period; and introducing corresponding weighting coefficients to obtain a comprehensive objective function as follows:
J＝w p J p +w e J e +w cx J cx +w cy J cy (7)
the decision-making behavior of the driver can be realized by calculating the objective function, the relevant vehicle state information which accords with the real driver is decided, and the information is output to a driver control behavior model;
s3, establishing a driver predicted behavior module
A three-degree-of-freedom vehicle model is adopted as an internal model of nonlinear model prediction control, and mainly describes transverse, longitudinal and yaw dynamic characteristics, and the three-degree-of-freedom vehicle model lists the following equations:
wherein m is the mass of the automobile, I z To the yaw moment of inertia, ω is the yaw rate of the vehicle, v x And v y Is the transverse and longitudinal speed, delta, of the vehicle in a coordinate system f And delta r For front and rear wheel steering angles, F lf And F sr Front wheel longitudinal tire force and front wheel lateral tire force, F, respectively lr And F sr Rear wheel longitudinal tire force and rear wheel lateral tire force, respectively;
the following vehicle dynamics formula is obtained after corresponding simplification and formula derivation:
wherein v is x And v y Is the transverse and longitudinal speed, F, of the vehicle in a coordinate system x Is the resultant force in the longitudinal direction of the vehicle tyre, a x As longitudinal acceleration of the vehicle, C f And C r Yaw stiffness, l, of the front and rear wheels of the vehicle, respectively f And l r Respectively the distance, delta, from the centre of mass of the car to the front and rear axes f Is the front wheel corner of the vehicle;
according to the basic principle of the nonlinear model predictive control, firstly, the vehicle-related state quantity at the future time is predicted according to the vehicle-related state at the current time, namely, the time domain N is predicted p Then, according to the related objective function, constraint and reference input, calculating out the corresponding control quantity, i.e. control time domain N c (ii) a Selecting the longitudinal speed, the lateral speed and the yaw rate of the vehicle as state quantities of the system, namely x = [ v = [ [ v ] x v y ω] T Selecting longitudinal acceleration and front wheel rotation angle as control quantity of system, i.e. u = [ a = x δ] T (ii) a The system can be expressed in the form of differential equations, i.e.
Discretizing the model, predicting the relevant state quantity of the vehicle at the future time, and taking a minimum sampling time T s And carrying out Euler dispersion on the system to obtain:
x(k+j+1)＝x(k+j)+T s g(x(k+j),u(k+j))
u(k+j)＝u(k+j-1)+Δu(k+j) (10)
through N p The state equation of the system obtained by predicting the steps is as follows:
x(k+1|k)＝x(k|k)+T s g(x(k|k),u(k|k))
x(k+2|k)＝x(k+1|k)+T s g(x(k+1|k),u(k+1|k))
x(k+N p |k)＝x(k+N p -1|k)+T s g(x(k+N p -1|k),u(k+N p -1|k)) (13)
predicting relevant variables at a future moment according to the state updating of the equation, namely the relevant state variables of the vehicle at the current moment;
s4, establishing a driver optimization behavior module
Simulating the optimization behavior of a driver in an objective function mode, solving the optimal control quantity, and establishing the following objective function:
wherein J 1 And J 2 Is an objective function for the tracking of the lateral and longitudinal speed of the vehicle, J 3 Is an objective function relating to the smoothness of the driver's operation, v x And v y Respectively, the predicted values of the lateral and longitudinal speeds of the vehicle, v x * And v y * Respectively as reference input of transverse and longitudinal speed of vehicle, delta u is variation of control input, Q 1 、Q 2 And Q 3 The weighting coefficients of the three objective functions respectively carry out corresponding terminal constraint on the control quantity,
solving the target function to obtain an optimal control sequence in a control time domain so as to minimize the value of the target function, taking a first value of the control sequence as output, and simultaneously converting a longitudinal acceleration signal into information of an accelerator pedal and a brake pedal of the vehicle;
s5, establishing a neuromuscular module of the driver
This phenomenon was simulated using the following transfer function:
wherein, t n Is the time constant of the neuromuscular response of the driver.
The invention has the beneficial effects that:
1. in the process of establishing the driver preview decision-making behavior model, the aspects of driving efficiency, smoothness in operation, stable driving and the like are considered, and meanwhile when a vehicle enters a curve, the vehicle speed can be reasonably decided according to the curvature of the road and the current vehicle speed, so that the behavior of the driver in curve deceleration is simulated.
2. In the process of establishing the driver control behavior model, the invention considers the transverse and longitudinal coupling behaviors of the driver, realizes the behaviors of steering, driving and braking of the driver on the vehicle in the driving process by utilizing the nonlinear model predictive control, and more effectively simulates the process of driving the vehicle by the real driver.
3. In the process of model building, the vehicle is regarded as a two-dimensional entity instead of simple particles, and the rectangle represents the safe driving outline of the vehicle, so that the method is more practical.
Drawings
FIG. 1 is a functional block diagram of a modeling method for driver cross-longitudinal coupling behavior;
FIG. 2 is a flow chart of driver preview decision behavior;
FIG. 3 is a schematic view of road boundary constraints;
FIG. 4 is a schematic view of curve driving;
FIG. 5 is a three-degree-of-freedom vehicle model diagram;
FIG. 6 is a schematic view of a vehicle travel track;
FIG. 7 is a schematic diagram of a simulated trajectory of a double-shift line segment and a rectangular outline of an automobile safety;
FIG. 8 is a graph of vehicle longitudinal speed change;
FIG. 9 is a graph of the change in longitudinal acceleration of the vehicle;
FIG. 10 is a graph of lateral acceleration change of a vehicle;
fig. 11 is a graph of steering wheel angle.
Detailed Description
When overtaking or changing lane, the speed of the vehicle needs to be changed properly to ensure the driving safety. Meanwhile, the longitudinal speed and the transverse speed of a real vehicle are not two completely decoupled variables, and a certain coupling relation exists between the longitudinal speed and the transverse speed, so that the invention provides a driver behavior model which can control the speed and the direction simultaneously on the basis of nonlinear model prediction control and by considering the coupling relation between the transverse behavior and the longitudinal behavior of a driver, and simultaneously, the process of driving pre-aiming decision is introduced, the curvature of a road is introduced, and the driver properly adjusts the vehicle speed according to the curvature of the road so as to ensure the safety of a curve.
The invention provides a modeling method for transverse and longitudinal coupling behaviors of a driver, which makes up the defects of the traditional driver model and is realized by the following steps:
step one, establishing a driver preview behavior module
During the driving of the vehicle, there is always a forward-looking process, which is called the driver's sighting behavior. According to the limitation of the automotive mobility and the current vehicle state information, feasible areas of the vehicle in the future preview time are calculated, the feasible areas simultaneously contain information such as the speed and the acceleration of the vehicle, and unsafe and unreasonable feasible areas are eliminated.
Step two, establishing a driver decision behavior module
And step one, obtaining a feasible region of the vehicle in the preview time, establishing an evaluation index, optimizing the track in the feasible thresholds, deciding a more ideal track, and outputting the related information such as the transverse and longitudinal speeds of the vehicle corresponding to the track.
Step three, establishing a driver predicted behavior module
The information obtained by the preview decision is used as the reference input of a driver prediction behavior model, and the state and position information of the vehicle at the next several moments are predicted according to the related information of the vehicle state at the moment, so that the degree of understanding of the driver on the performance of the vehicle is reflected.
Step four, establishing a driver optimization behavior module
After the preview information and the prediction information are obtained, in order to make the prediction information of the driver and the preview information of the driver tend to be consistent, the driver needs to find out the optimal control behaviors, namely the longitudinal acceleration and the steering wheel rotation angle of the vehicle, so that the vehicle runs according to the state decided by the preview of the driver, and the longitudinal acceleration signal is converted into driving and braking information so as to control the controlled vehicle.
Step five, establishing a neuromuscular module of the driver
The driving or braking information and the steering wheel angle information obtained through the above steps need to be transmitted to the controlled vehicle through the neuromuscular of the driver, so that a certain response delay exists, generally, the delay belongs to a pure delay, and the delay behavior of the driver is expressed by using a transfer function.
The present invention is described in detail below with reference to fig. 1 to 11:
the driver transverse and longitudinal coupling behavior model is mainly realized by a driver aiming behavior module, a driver decision behavior module, a driver forecasting behavior module, a driver optimizing behavior module and a driver neuromuscular module, and comprises the processes of aiming, decision making and execution when a driver drives a vehicle.
The driver pre-aiming behavior module is mainly used for describing that the sight line of a driver is always in the process of looking forward in the process of driving the automobile, observing the feasible region of a road ahead, and giving a prerequisite for a subsequent driver to judge how to control the automobile; the driver decision behavior module is mainly used for describing, deciding the safe and legal optimal vehicle running track in the feasible region through the related evaluation indexes, and outputting the vehicle related state information corresponding to the track; the driver prediction behavior module mainly describes the prediction behavior of the driver on the vehicle state and the position at the following moments according to the vehicle state at the moment, and reflects the cognitive condition of the driver on the performance of the vehicle; the driver optimization behavior module is mainly used for describing the process of finding the optimal vehicle control behavior by a driver, so that the vehicle can run in a mode decided by a preview, and simultaneously, longitudinal acceleration signals output by the model are converted into driving and braking signals capable of acting on the controlled vehicle; the driver neuromuscular module is mainly used to describe delayed behavior on the neuromuscular that the driver presents itself.
The specific steps of the embodiment are as follows:
a basic principle block diagram of a driver model is shown in fig. 1, wherein a preview decision-making behavior module of a driver is mainly used for simulating a preview behavior of the driver when the driver drives a car and making a decision of relevant preview information, and the preview information comprises a maneuvering capability limitation module, a motion trajectory prediction module, a safety constraint module, a legality constraint module and a comprehensive performance evaluation module; the control behavior module of the driver is used for simulating the operation behavior of the driver for driving the automobile after the preview decision, and comprises a prediction behavior, an optimization behavior, a neuromuscular delay module and a controlled vehicle; the method comprises the following steps.
Step one, establishing a driver preview behavior module
As shown in fig. 2, the flow chart of the preview decision-making behavior of the driver shows that the automobile has corresponding mobility limitation, and the maximum acceleration and deceleration that can be achieved by the automobile has a certain range, that is, the maximum acceleration and deceleration that can be achieved by the automobile is within a certain range
Wherein A is x And A y Respectively the longitudinal and lateral acceleration of the vehicle. Respectively carrying out appropriate discretization on the transverse acceleration and the longitudinal acceleration to form a two-dimensional plane, wherein each node on the two-dimensional plane corresponds to a group of transverse and longitudinal accelerations, and the distance between each discrete point of the transverse and longitudinal accelerations is P x ，P y The motion trail of the vehicle can be predicted at each group of the transverse and longitudinal accelerations, and the predicted trail also comprises other vehicle state related information, such as the transverse and longitudinal speeds, the acceleration, the vehicle position information and the like of the vehicle. And after the corresponding predicted track information is obtained, judging the safety and the legality of each track, deleting unsafe and illegal tracks, and deciding the optimal transverse and longitudinal acceleration, namely the point Q in the graph through comprehensive evaluation indexes. Then judging P at this time x ，P y And if the required precision is not met, re-discretizing the area near the currently optimal longitudinal and transverse acceleration point Q, and repeating the process again until the precision is met.
After the preliminary discretization of the transverse and longitudinal accelerations, the preview time T is carried out on the transverse and longitudinal accelerations of each group p The prediction of the motion track is that the aiming time is divided into m equal parts
And assuming that the transverse and longitudinal acceleration and the centroid slip angle of the vehicle in the prealignment time are fixed values, and utilizing the following formula to carry out vehicle state information in the prealignment time:
and updating the position information of the vehicle using the following formula:
wherein X and Y are the horizontal and vertical coordinates of the mass center of the vehicle in the geodetic coordinate system, v x And v y Is the transverse and longitudinal speed of the vehicle in the vehicle coordinate system, A x And A y Is the transverse and longitudinal acceleration of the vehicle,
the heading angle of the vehicle, omega is the yaw angular velocity of the vehicle, beta is the centroid slip angle of the vehicle, and j =1,2,3 \8230m.
According to the position information and the state information of the vehicle at the current moment, the motion trail prediction in the preview time can be carried out on each group of discretized transverse and longitudinal acceleration through the formula, and meanwhile, the motion trails also contain the related state information of the vehicle. The driving feasible region of the current vehicle can be obtained through the motion track prediction, and unsafe and illegal tracks in the feasible region need to be deleted.
Firstly, safety constraint is carried out on the road safety vehicle, and the road safety vehicle is mainly divided into two parts, namely road boundary safety constraint and curve safety vehicle speed constraint. When the vehicle runs on the road, the vehicle body keeps a certain distance from the road boundary or other obstacles in order that the driver does not collide with the road boundary or other obstacles, and a rectangle is used to represent the safety distance and is called as a safety rectangle of the vehicle. As shown in fig. 3, since there is a part of the vehicle feasible region predicted from the motion trail, which may cause the vehicle to run out of the road boundary, resulting in a collision, when the road boundary safety constraint is performed, the feasible regions should be deleted to ensure the driving safety. And calculating the coordinates of four vertexes of the safety rectangle according to the coordinates of the mass center of the vehicle and the course angle by using the following formula:
wherein X A And Y A Is the horizontal and vertical coordinates of the safe rectangle vertex A, and the like, X and Y are the horizontal and vertical coordinates of the vehicle mass center respectively,
l and W are the vehicle heading angle, and the length and width of the vehicle safety rectangle. Through the coordinates of the four vertexes and the coordinates of the center of mass of the safety rectangle, whether the vehicle collides with the road boundary can be converted into the position relation between the rectangle and the simple polygon, if the vehicle does not collide with the road boundary, the vehicle is kept, otherwise, the motion trail is deleted.
When the vehicle runs to a curve, the speed of the vehicle is critical, and when the speed of the vehicle is too high, a real driver can perform a deceleration action in front so that the vehicle can safely pass through the curve. The curve-safe vehicle speed constraint thus reflects mainly the behavior of the driver to adjust the vehicle speed to within the safe vehicle speed in advance when entering the curve. As shown in fig. 4, at the time when the vehicle travels at a curve, there is the following relationship:
where F is the lateral force of the vehicle at the curve, V is the driving speed of the vehicle, R is the curvature radius of the road, μ is the friction coefficient of the road, and θ is the inclination angle of the road. The turning radius of the vehicle approximates the curvature radius of the road. In order to prevent the vehicle from rolling over when the vehicle runs on a curve, the basic condition that the vehicle safely runs through the curve is met. Meanwhile, different drivers can drive at different speeds on the premise of ensuring safety.
When the road inclination angle is zero, the speed safety constraint of the vehicle in the curve can be obtained by introducing the style coefficient epsilon:
therefore, in the driving process of the vehicle, the maximum safe speed of the vehicle in the curve which is considered by the driver to drive can be obtained through calculating the real-time road curvature, and the vehicle speed can be correspondingly adjusted, so that the vehicle can safely pass through the curve. The two safety constraints make the driver model more consistent with real driver characteristics.
After safety constraint, the legality of the vehicle is also constrained, and the vehicle speed is limited by regulations mainly, namely the maximum longitudinal speed of the vehicle should not exceed the speed limit by the regulations. If the vehicle speed exceeds the regulation speed limit in the preview time, deleting the motion trail in the feasible domain to realize the legality constraint.
Step two, establishing a driver decision behavior module
The feasible region information obtained through the preview model of the driver still has a plurality of vehicle feasible driving tracks, and the relevant information conforming to the real driver can be output only through decision making. The method mainly makes a decision on the feasible region according to three aspects of driving stability, driving efficiency and operation smoothness. The driving stability mainly describes the behavior of a driver to prevent the vehicle from shaking left and swinging right in the driving process and driving close to the center line of a road in order to ensure the stability of the driving track of the vehicle; the driving efficiency is mainly described, a driver completes the behavior of a driving task at a higher speed on the premise of ensuring the safety and the legality, and the driving efficiency is higher as the speed is higher; the smoothness of operation mainly describes the behavior of the driver without making a large abrupt operation of the vehicle. According to the three aspects, the following objective functions are respectively established:
wherein, J p An objective function representing the driving stationarity, d j The distance from the mass center of the vehicle to the center line of the road at the jth point in the motion trail prediction is obtained; j. the design is a square e An objective function representing driving efficiency, A x Is the vehicle longitudinal acceleration; j. the design is a square cx And J cy Representing smoothness of driving operation, Δ A x And Δ A y The change of the lateral acceleration and the longitudinal acceleration is represented, namely the lateral acceleration and the longitudinal acceleration change value of the decision period is larger than that of the previous decision period.
The denominators in the objective function are all corresponding standard values which are fixed constants. The comprehensive objective function can be obtained by introducing corresponding weighting coefficients as follows:
J＝w p J p +w e J e +w cx J cx +w cy J cy (7)
according to the objective function, the smaller the value is, the more the value is consistent with the real driver behavior, the decision-making behavior of the driver can be realized through calculation of the objective function, the relevant vehicle state information consistent with the real driver is decided, and the information is output to the driver control behavior model.
Step three, establishing a driver predicted behavior module
The driver prediction behavior model mainly adopts a nonlinear model prediction control method, and because the driver model needs to control the transverse and longitudinal directions of the vehicle at the same time, a three-degree-of-freedom vehicle model is adopted as an internal model of the nonlinear model prediction control, mainly describing transverse, longitudinal and yaw dynamics characteristics, and meeting the basic requirements of the driver model, as shown in fig. 5, the schematic diagram of the three-degree-of-freedom vehicle model can list the following equations according to the diagram:
wherein m is the mass of the automobile, I z Is the yaw moment of inertia, ω is the yaw rate of the vehicle, v x And v y Is the transverse and longitudinal speed, delta, of the vehicle in a coordinate system f And delta r For front and rear wheel steering angles, F lf And F sr Front wheel longitudinal tire force and front wheel lateral tire force, F, respectively lr And F sr Respectively, the rear wheel longitudinal tire force and the rear wheel lateral tire force.
In order to make the model simple and convenient and meet the required requirements, the following vehicle dynamics formula is obtained through corresponding simplification and formula derivation:
wherein v is x And v y Is the transverse and longitudinal speed, F, of the vehicle in a coordinate system x Is the resultant force in the longitudinal direction of the vehicle tyre, a x As longitudinal acceleration of the vehicle, C f And C r Are respectively front and rear wheels of the automobileLateral deflection stiffness of l f And l r Respectively, the distance from the center of mass of the vehicle to the front and rear axes, delta f Is the front wheel angle of the vehicle.
According to the basic principle of the nonlinear model predictive control, firstly, the vehicle-related state quantity at the future time is predicted according to the vehicle-related state at the current time, namely, the time domain N is predicted p Then, according to the related objective function, constraint and reference input, calculating out the corresponding control quantity, i.e. control time domain N c . Selecting the longitudinal speed, the lateral speed and the yaw rate of the vehicle as state quantities of the system, namely x = [ v = [ [ v ] x v y ω] T Selecting longitudinal acceleration and front wheel rotation angle as control quantity of system, i.e. u = [ a = x δ] T . The system can be expressed in the form of differential equations, i.e.
Carrying out appropriate discretization on the model, predicting the relevant state quantity of the vehicle at the future time, and taking a minimum sampling time T s And carrying out Euler dispersion on the system to obtain:
through N p The state equation of the system obtained after the prediction of the steps is as follows:
through the state updating of the formula, the relevant variables at the future time can be predicted according to the relevant state variables of the vehicle at the current time, and the understanding of the driver on the performance of the vehicle is reflected.
Step four, establishing a driver optimization behavior module
By obtaining the predicted value of the relevant state quantity in the prediction time domain, the driver always wants to exchange the operation quantity as small as possible for the optimal operation effect in the process of the driver, namely the actual motion information of the vehicle accords with the predicted relevant information. Therefore, the optimization behavior of the driver is simulated in an objective function mode, the optimal control quantity is solved, and the following objective function is established:
wherein J 1 And J 2 Is an objective function for the tracking of the lateral and longitudinal speed of the vehicle, J 3 Is an objective function, v, relating to the smoothness of the driver's operation x And v y Respectively, the predicted values of the lateral and longitudinal speeds of the vehicle, v x * And v y * Reference input for the lateral and longitudinal speed of the vehicle, respectively, and Δ u the variation of the control input, i.e. the variation of the longitudinal acceleration and the steering wheel angle, Q 1 、Q 2 And Q 3 Respectively, the weighting coefficients of the three objective functions.
In order to make the vehicle run smoothly, corresponding terminal constraint needs to be carried out on the control quantity, namely:
in order to enable the control effect of the vehicle to reach the result of the preview decision of the driver, the objective function is solved to obtain an optimal control sequence in a control time domain so as to enable the value of the objective function to be minimum, the first value of the control sequence is used as output, and meanwhile, the longitudinal acceleration signal is converted into information of an accelerator pedal and a brake pedal of the vehicle, so that the controlled vehicle can be controlled in the subsequent process.
Step five, establishing a neuromuscular module of the driver
The neuromuscular model of the driver is mainly used for representing that when the driver performs the expected operation action, certain physiological limits exist, generally, the delay can be regarded as a pure lag, and therefore, the phenomenon is simulated by using the following transfer function:
wherein, t n The time constant of the neuromuscular response of the driver.
Joint simulation verification
The five steps are built in simulink in matlab, and B-Class in Carsim is selected as the controlled vehicle, and the relevant parameters of the vehicle are shown in Table 1:
TABLE 1
In order to form a closed loop between Carsim and the driver model, corresponding input and output ports are required to be arranged on the Carsim, wherein the input port comprises steering wheel turning angle, throttle control and brake control, and the output port comprises vehicle transverse and longitudinal speed, yaw angular speed, course angle, engine speed and gear information. The simultaneous setting of the relevant parameters of the driver model is shown in Table 2
TABLE 2
Through the setting of the relevant parameters, the driver model and the controlled vehicle form a closed loop, the model is subjected to simulation verification by adopting a double-shift working condition, as shown in fig. 6, the model is a comparison graph of the driving track of the vehicle and the central line of the road, and as can be seen from the graph, the driver model can better simulate the steering behavior of the driver. As shown in fig. 7, the diagram is a schematic diagram of a simulated trajectory of a double-moving route section and a contour of a safety rectangle of an automobile, in the diagram, a ratio of a horizontal axis to a vertical axis is greatly different, but the size of the solid-line rectangle is an actual safety contour rectangle of the automobile, and the whole driving process does not collide with a road boundary, so that the driving safety is met. As shown in fig. 8, it can be known from the road condition information that the longitudinal variation curve of the vehicle in the vehicle coordinate system is a longitudinal variation curve of the vehicle, the vehicle is in an acceleration state when it starts to be in a straight-driving stage, the vehicle decelerates and passes through a curve at a safe speed when it reaches the first curve, the speed of the vehicle is adjusted according to the curvature of the road at the lower curve, and finally the vehicle enters the straight-driving stage and accelerates to the maximum speed, so that the acceleration behavior is stopped. As shown in fig. 9, 10, 11, there are a vehicle longitudinal acceleration chart, a vehicle lateral acceleration chart, and a steering wheel angle chart, respectively. The actual validity of the driver model can be verified through the joint simulation.
Description of the symbols involved in the invention:
A x : longitudinal acceleration of vehicle
A y : lateral acceleration of vehicle
P x 、P y : dispersion of longitudinal and lateral acceleration
T p : preview time
t: discrete time after equal division of preview time m
j time sequence after equal division of preview time
ω j Yaw rate of vehicle at time j
Beta: vehicle centroid slip angle
X j 、X j-1 : abscissa of vehicle mass center at j-1 and j time under geodetic coordinate system
Y j 、Y j-1 Ordinate of vehicle mass center at time j-1 and time j in geodetic coordinate system
X A 、X B 、X C 、X D : abscissa of four vertexes of vehicle safety rectangle
Y A 、Y B 、Y C 、Y D : ordinate of four vertexes of vehicle safety rectangle
X, Y: horizontal and vertical coordinates of vehicle mass center
L, W: length and width of vehicle safety rectangle
F: lateral force of vehicle at curve
V: running speed of vehicle
R: radius of curvature of road
μ: coefficient of friction of road
θ: inclination angle of road
Epsilon: curve speed style coefficient
V max : maximum safe speed of curve
J p : objective function of driving stability
d j : distance from jth point vehicle mass center to road center line in motion trail prediction
J e : target function of driving efficiency
J cx 、J cy : smooth target function of driving operation
ΔA x 、ΔA y : variation of transverse and longitudinal acceleration
w p 、w e 、w cx 、w cy : weight coefficient corresponding to objective function
J: synthetic objective function
m: vehicle mass
I z : yaw moment of inertia
ω: yaw rate of vehicle
δ f 、δ r : front wheel steering angle and rear wheel steering angle
F lf 、F sf : front wheel longitudinal and lateral tire forces
F lr 、F sr : rear wheel longitudinal and lateral tire forces
F x : resultant force in longitudinal direction of vehicle tyre
a x : longitudinal acceleration of vehicle
C f 、C r : cornering stiffness of front and rear wheels of a vehicle
l f 、l r : distance from automobile mass center to front and rear axes
N p : predicting the time domain
N c : controlling time domain
x: variable of state
u: controlled variable
δ: steering wheel corner
T s : sampling time
g (x (t), u (t)): system equation of state
J 1 、J 2 : target function for tracking transverse and longitudinal speed of vehicle
J 3 : target function of driver's ride comfort
v x (k+j|k)、v y (k + j | k): predicted values of longitudinal and lateral vehicle speeds
v x * 、v y * : reference input for lateral and longitudinal speed of vehicle
Δ u: variation of control input
Q 1 、Q 2 、Q 3 ：J 1 、J 2 、J 3 Weighting coefficient of
u max : controlling the maximum value of the input quantity
δ max : maximum value of steering wheel angle
a x : maximum value of longitudinal acceleration of vehicle
t n : time constant of the neuromuscular response of the driver.
Claims (1)
1. A driver transverse and longitudinal coupling behavior model is characterized in that:
s1, establishing a driver preview behavior module
After the preliminary discretization of the transverse and longitudinal accelerations, the preview time T is carried out on the transverse and longitudinal accelerations of each group p The prediction of the motion track is that the aiming time is divided into m equal parts
And assuming that the transverse and longitudinal acceleration and the centroid slip angle of the vehicle in the prealignment time are fixed values, and utilizing the following formula to carry out vehicle state information in the prealignment time:
and updating the position information of the vehicle using the following formula:
wherein X and Y are the horizontal and vertical coordinates of the mass center of the vehicle in the geodetic coordinate system, v x And v y Is the transverse and longitudinal speed of the vehicle in the vehicle coordinate system, A x And A y Is the transverse and longitudinal acceleration of the vehicle,
the heading angle of the vehicle, omega is the yaw angular velocity of the vehicle, beta is the mass center side slip angle of the vehicle, and j =1,2,3 \8230m;
the safety constraint is divided into two parts, namely road boundary safety constraint and curve safety vehicle speed constraint;
and calculating the coordinates of four vertexes of the safety rectangle according to the coordinates of the mass center of the vehicle and the course angle by using the following formula:
wherein X A And Y A Is the horizontal and vertical coordinates of the safe rectangle vertex A, and the like, X and Y are respectively the horizontal and vertical coordinates of the vehicle mass center,
l and W are the length and width of the vehicle safety rectangle; whether the vehicle collides with the road boundary or not can be converted into the position relation between the rectangle and the simple polygon through the coordinates of the four vertexes and the coordinates of the mass center of the safety rectangle, if the vehicle does not collide with the road boundary, the vehicle is kept, otherwise, the motion trail is deleted;
and (3) safety vehicle speed restraint at the bend: at a curve where the vehicle is traveling, the following relationship exists:
wherein F is the lateral force of the vehicle at the curve, V is the running speed of the vehicle, R is the curvature radius of the road, mu is the friction coefficient of the road, and theta is the inclination angle of the road;
when the road inclination angle is zero, the speed safety constraint of the vehicle in the curve can be obtained by introducing the style coefficient epsilon:
in the driving process of the vehicle, the maximum safe speed of the vehicle in the curve considered by a driver during driving can be obtained by calculating the real-time road curvature, and the vehicle speed can be correspondingly adjusted to ensure that the vehicle safely passes through the curve;
s2, establishing a driver decision behavior module
Making a decision on a feasible region according to three aspects of driving stability, driving efficiency and operation smoothness, and respectively establishing the following objective functions:
wherein, J p An objective function representing the driving stationarity, d j The distance from the mass center of the vehicle to the center line of the road at the jth point in the motion trail prediction is obtained; j is a unit of e An objective function representing driving efficiency, A x Is the vehicle longitudinal acceleration; j. the design is a square cx And J cy Representing smoothness of driving operation, Δ A x And Δ A y Representing the variation of the horizontal and vertical acceleration, namely the horizontal and vertical acceleration variation value of the decision period is larger than that of the previous decision period; the comprehensive objective function is obtained by introducing corresponding weighting coefficients as follows:
J＝w p J p +w e J e +w cx J cx +w cy J cy (7)
the decision-making behavior of the driver can be realized by calculating the objective function, the relevant vehicle state information which accords with the real driver is decided, and the information is output to a driver control behavior model;
s3, establishing a driver predicted behavior module
A three-degree-of-freedom vehicle model is adopted as an internal model of nonlinear model prediction control, and mainly describes transverse, longitudinal and yaw dynamic characteristics, and the three-degree-of-freedom vehicle model lists the following equations:
wherein m is the mass of the automobile, I z To the yaw moment of inertia, ω is the yaw rate of the vehicle, v x And v y Is the transverse and longitudinal speed, delta, of the vehicle in a coordinate system f And delta r For front and rear wheel steering angles, F lf And F sr Front wheel longitudinal tire force and front wheel lateral tire force, F, respectively lr And F sr Rear wheel longitudinal tire force and rear wheel lateral tire force, respectively;
the following vehicle dynamics formula is obtained after corresponding simplification and formula derivation:
wherein v is x And v y Is the transverse and longitudinal speed, F, of the vehicle in a coordinate system x Is the resultant force in the longitudinal direction of the vehicle tyre, a x As longitudinal acceleration of the vehicle, C f And C r Yaw stiffness, l, of the front and rear wheels of the vehicle, respectively f And l r Respectively, the distance from the center of mass of the vehicle to the front and rear axes, delta f Is the front wheel corner of the vehicle;
according to the basic principle of nonlinear model predictive control, firstly, the vehicle-related state quantity at the future time is predicted according to the vehicle-related state at the current time, namely, the time domain N is predicted p Then, according to the related objective function, constraint and reference input, calculating out the corresponding control quantity, i.e. control time domain N c (ii) a Selecting the longitudinal speed, the lateral speed and the yaw rate of the vehicle as state quantities of the system, namely x = [ v = x v y ω] T SelectingTaking the longitudinal acceleration and the front wheel rotation angle as the control quantity of the system, i.e. u = [ a ] x δ] T (ii) a The system can be expressed in the form of differential equations, i.e.
Discretizing the model, predicting the relevant state quantity of the vehicle at the future time, and taking a minimum sampling time T s And carrying out Euler dispersion on the system to obtain:
x(k+j+1)＝x(k+j)+T s g(x(k+j),u(k+j))
u(k+j)＝u(k+j-1)+Δu(k+j) (10)
through N p The state equation of the system obtained by predicting the steps is as follows:
predicting relevant variables at a future moment according to the state updating of the formula, namely the relevant state variables of the vehicle at the current moment;
s4, establishing a driver optimization behavior module
Simulating the optimization behavior of a driver in an objective function mode, solving the optimal control quantity, and establishing the following objective function:
wherein J 1 And J 2 Is an objective function for the tracking of the lateral and longitudinal speeds of the vehicle, J 3 Is an objective function relating to the smoothness of the driver's operation, v x And v y Respectively, the predicted values of the lateral and longitudinal speeds of the vehicle, v x * And v y * Respectively as reference input of transverse and longitudinal speed of vehicle, delta u as variation of control input, Q 1 、Q 2 And Q 3 The weighting coefficients of the three objective functions respectively carry out corresponding terminal constraint on the control quantity,
solving the target function to obtain an optimal control sequence in a control time domain so as to minimize the value of the target function, taking a first value of the control sequence as output, and simultaneously converting a longitudinal acceleration signal into information of an accelerator pedal and a brake pedal of the vehicle;
s5, establishing a neuromuscular module of the driver
This phenomenon was simulated using the following transfer function:
wherein, t n Is the time constant of the neuromuscular response of the driver.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211054892.4A CN115292671A (en) | 2022-08-31 | 2022-08-31 | Driver horizontal-vertical coupling behavior model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211054892.4A CN115292671A (en) | 2022-08-31 | 2022-08-31 | Driver horizontal-vertical coupling behavior model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115292671A true CN115292671A (en) | 2022-11-04 |
Family
ID=83831152
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211054892.4A Pending CN115292671A (en) | 2022-08-31 | 2022-08-31 | Driver horizontal-vertical coupling behavior model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115292671A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114906173A (en) * | 2022-06-30 | 2022-08-16 | 电子科技大学 | Automatic driving decision-making method based on two-point preview driver model |
-
2022
- 2022-08-31 CN CN202211054892.4A patent/CN115292671A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114906173A (en) * | 2022-06-30 | 2022-08-16 | 电子科技大学 | Automatic driving decision-making method based on two-point preview driver model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111338340B (en) | Model prediction-based local path planning method for unmanned vehicle | |
| CN107315411B (en) | Lane changing track planning method for unmanned vehicle based on vehicle-vehicle cooperation | |
| CN107380161B (en) | A kind of active steering control device for aiding in driver to realize desired ride track | |
| CN108819951B (en) | Man-machine common driving transverse driving right distribution method considering driver driving skill | |
| CN112622886B (en) | Anti-collision early warning method for heavy operation vehicle comprehensively considering front and rear obstacles | |
| CN109606368B (en) | Intelligent automobile extension vehicle speed self-adaptive change trajectory tracking control method | |
| CN112046484B (en) | Q learning-based vehicle lane-changing overtaking path planning method | |
| CN110161865B (en) | Intelligent vehicle track change trajectory planning method based on nonlinear model predictive control | |
| CN108569336A (en) | Vehicle kinematics model rotating direction control method is based under Dynamic Constraints | |
| CN104881030A (en) | Unmanned vehicle-side longitudinal coupling tracking control method based on rapid terminal sliding mode principle | |
| CN110091868A (en) | A kind of longitudinal collision avoidance method and its system, intelligent automobile of man-machine coordination control | |
| CN112578672B (en) | Unmanned vehicle trajectory control system based on chassis nonlinearity and trajectory control method thereof | |
| CN112896188B (en) | Automatic driving decision control system considering front vehicle encounter | |
| CN108897216A (en) | A kind of intelligent automobile model prediction decision-making technique based on the virtual field of force in region | |
| CN113715842A (en) | High-speed moving vehicle control method based on simulation learning and reinforcement learning | |
| CN115292671A (en) | Driver horizontal-vertical coupling behavior model | |
| CN114987461A (en) | Intelligent passenger car dynamic lane change trajectory planning method under multi-car complex traffic environment | |
| CN107544496A (en) | A kind of method for planning track for improving cosine lane-change model | |
| CN115422837A (en) | Intelligent automobile dynamics prediction model based on deep Gaussian process, training data acquisition method and training method | |
| CN114889589A (en) | Intelligent automobile steering and braking cooperative collision avoidance control system and method | |
| CN107992039B (en) | Trajectory planning method based on flow field in dynamic environment | |
| CN115042770A (en) | Vehicle queue transverse control method based on distributed robust model prediction | |
| CN114043984A (en) | Intelligent automobile lane change control system and method based on Internet of vehicles environment | |
| Liu et al. | Research on Vehicle Lane Change Based on Vehicle Speed Planning | |
| Choi et al. | Coordinated steering angle and yaw moment distribution to increase vehicle regenerative energy in autonomous driving |
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 |