CN107885932B - Automobile emergency collision avoidance layered control method considering man-machine harmony - Google Patents

Automobile emergency collision avoidance layered control method considering man-machine harmony Download PDF

Info

Publication number
CN107885932B
CN107885932B CN201711082166.2A CN201711082166A CN107885932B CN 107885932 B CN107885932 B CN 107885932B CN 201711082166 A CN201711082166 A CN 201711082166A CN 107885932 B CN107885932 B CN 107885932B
Authority
CN
China
Prior art keywords
automobile
reference value
path
moment
control
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.)
Expired - Fee Related
Application number
CN201711082166.2A
Other languages
Chinese (zh)
Other versions
CN107885932A (en
Inventor
李绍松
郭陆平
于志新
崔高健
郑顺航
朱峰挺
闫旭
刘秀峰
吴晓东
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.)
Changchun University of Technology
Original Assignee
Changchun University of Technology
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 Changchun University of Technology filed Critical Changchun University of Technology
Priority to CN201711082166.2A priority Critical patent/CN107885932B/en
Publication of CN107885932A publication Critical patent/CN107885932A/en
Application granted granted Critical
Publication of CN107885932B publication Critical patent/CN107885932B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D5/00—Power-assisted or power-driven steering
    • B62D5/04—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear
    • 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
    • B60W2710/00—Output or target parameters relating to a particular sub-units
    • B60W2710/20—Steering systems
    • B60W2710/202—Steering torque

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Combustion & Propulsion (AREA)
  • Computational Mathematics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Peptides Or Proteins (AREA)
  • Steering Control In Accordance With Driving Conditions (AREA)

Abstract

A layered control method for automobile emergency collision avoidance considering man-machine harmony relates to the technical field of automobile auxiliary driving, and the method optimizes in real time to obtain a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track through a path dynamic planning module according to barrier information, target point coordinates and automobile driving state information which are collected in real time, inputs the lateral displacement reference value, the yaw angle speed reference value and the longitudinal speed reference value into a path tracking control module, collects current automobile driving state information through the path tracking control module, optimizes in real time to obtain front wheel corners and four wheel slip rates of an automobile, and controls the automobile to realize collision avoidance; in the collision avoidance control process, the EPS moment compensation module determines the compensation control moment according to the vehicle speed and the additional rotation angle of the front wheel, and controls the sudden change moment of the steering wheel within an ideal range, so that the man-machine harmonious emergency collision avoidance of the automobile is realized. The invention solves the problems of dynamic path planning and real-time tracking during emergency collision avoidance and realizes safe and optimal collision avoidance.

Description

Automobile emergency collision avoidance layered control method considering man-machine harmony
Technical Field
The invention relates to the technical field of automobile auxiliary driving, in particular to an automobile emergency collision avoidance layered control method considering man-machine harmony.
Background
The automobile can bring convenience and quickness to people, and the driving safety of the automobile becomes a global social problem. In order to further improve road traffic safety and help drivers to reduce erroneous operations, attention has been paid to and intelligent automobile safety technologies represented by Advanced Driver Assistance Systems (ADAS) in recent years. The automobile emergency collision avoidance system assists a driver to adjust the motion track of an automobile through active intervention of an actuator, so that collision avoidance is realized. The novel bicycle can save lives of drivers at critical moment, and has good market prospect.
Real-time planning and tracking of a collision-free optimal path are the key points of automobile emergency collision avoidance control. The collision avoidance control of the automobile needs the automobile to continuously plan an expected path on the premise of acquiring automobile state information and road information, and simultaneously assists a driver to complete optimal steering and braking operations, so that the safe collision avoidance of the automobile is realized. Because the driving track of the automobile and the corresponding Control input need to be optimized in real time, in recent years, with the breakthrough of a Model Predictive Control (MPC) theory based on real-time mathematical optimization, the MPC has been rapidly expanded from the slow process industries such as chemical engineering to the fast Control systems such as aerospace, robots, automobiles, and the like. However, in emergency collision avoidance, due to the complexity of the model, it is difficult for the automobile to meet the real-time requirement on the premise of ensuring accurate control, which is always a main factor limiting the application of MPC.
The active intervention of the steering system is not left in the automobile emergency collision avoidance control. European regulations require that there be a mechanical connection between the Steering wheel and the steered wheels, so Active Front Steering (AFS) has come into force as a transition product to the steer-by-wire (SBW) in the future. AFS, while changing the system displacement transmission characteristics, also affects the force transmission characteristics of the steering system, causing abrupt changes in the steering wheel Torque, see document 1[ Sumioaugita, Masayoshi Tomizka. calibration of Unnatual Reaction Torque in variable-wheel-Ratio [ J ]. Journal of Dynamic Systems Measurement & Control,2012,134(2):021019.A ] and document 2[ tsushi Oshima, Xu Chen, Sumio Sugita, Masayoshi Tomizuka. Control design for calibration of environmental Reaction and vibration in variable-Gear-Ratio system [ C ]. 2013Dynamic Control of vibration system, parameter of 20111. model 3797, sample 3711. V.S.A.: 1. A. The excessive sudden change moment of the steering wheel can aggravate the nervous mind of a driver, so that the driver is easy to operate by mistake, and the driving safety is not facilitated. The proper abrupt change moment of the steering wheel is beneficial to the driver to sense the attitude change of the automobile and plays a role in warning. The driver's acceptance of the steering wheel snap torque varies from person to person.
Disclosure of Invention
In order to solve the technical problem that the sudden change of torque of a steering wheel is uncontrollable and misoperation of a driver is easily caused in the conventional emergency collision avoidance method, the invention provides a layered control method for automobile emergency collision avoidance considering man-machine harmony, which can assist the driver to finish collision avoidance and save lives of drivers and passengers at emergency shut-off.
The technical scheme adopted by the invention for solving the technical problem is as follows:
an automobile emergency collision avoidance layered control method considering man-machine harmony is disclosed, which comprises the following steps: the method comprises the steps that a path dynamic planning module optimizes in real time to obtain a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track according to barrier information, target point coordinates and automobile running state information which are collected in real time, the lateral displacement reference value, the yaw angle speed reference value and the longitudinal speed reference value are input into a path tracking control module, meanwhile, the path tracking control module collects current automobile running state information, front wheel corners and automobile four-wheel slip rates are obtained through real-time optimization, and the automobile is controlled to avoid collision; in the process of controlling collision avoidance, a compensation control moment is determined by an Electric Power Steering (EPS) moment compensation module according to the vehicle speed and the additional rotation angle of the front wheel, and the sudden change moment of a Steering wheel is controlled in an ideal range, so that man-machine harmonious automobile emergency collision avoidance is realized; the method comprises the following steps:
step 1, a path dynamic planning module optimizes in real time to obtain a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track according to barrier information, target point coordinates and automobile running state information which are collected in real time, and the path dynamic planning module comprises the following substeps:
step 1.1, the performance index design process of the dynamic path planning comprises the following substeps:
step 1.1.1, using a two-norm of the error between the terminal point coordinate of the predicted track in the predicted time domain and the target point coordinate as a tracking performance index to reflect the track tracking characteristic of the automobile, wherein the expression is as follows:
Figure BDA0001459254100000021
wherein Hp，hFor the prediction time domain of the dynamic path planning module, (X)t+Hp，h,Yt+Hp，h) The coordinates (X) of the target point to be reached by the automobile in collision avoidance are obtained by iteration of the particle model for predicting the terminal point coordinates of the predicted track in the time domaing,Yg)；
The particle model is:
Figure BDA0001459254100000022
Figure BDA0001459254100000023
Figure BDA0001459254100000024
Figure BDA0001459254100000025
Figure BDA0001459254100000026
wherein the content of the first and second substances,
Figure BDA0001459254100000027
ayis the lateral acceleration speed of the automobile;
Figure BDA0001459254100000028
is the longitudinal acceleration of the vehicle;
Figure BDA0001459254100000029
respectively the automobile yaw angle and the yaw angular speed;
Figure BDA00014592541000000210
respectively the longitudinal speed and the lateral speed of the mass center of the automobile in a geodetic coordinate system; v is the current longitudinal speed of the automobile;
step 1.1.2, utilizing the two-norm of the lateral acceleration as the automobile safety index in the collision avoidance process to embody the automobile collision avoidance stability, and establishing the discrete quadratic automobile safety index as follows:
Figure BDA0001459254100000031
wherein Hc，hA control time domain of a dynamic path planning module, t represents the current moment, ayIs the lateral acceleration, w, of the particle model1Is ayThe weight coefficient of (a);
step 1.2, the constraint design process of the dynamic path planning comprises the following substeps:
step 1.2.1, setting automobile stability constraint to ensure the safety of automobile obstacle avoidance;
and (3) limiting the upper limit and the lower limit of the lateral acceleration by using a linear inequality to obtain the stability constraint of the automobile, wherein the mathematical expression is as follows:
|ayk，t|＜μg k＝t，t+1……t+Hc，h-1 (3)
wherein mu is a road surface adhesion coefficient, and g is a gravity acceleration;
step 1.2.2, setting position constraint to ensure that the collision with an obstacle is avoided in the collision avoidance process;
the position information of the obstacle at time t can be characterized as a set of N discrete points, which can be measured by a radar sensor, wherein the coordinate of the jth discrete point is expressed as (X)j,t,Yj,t) And the coordinate of the mass center of the automobile at the moment t is recorded as (X)k,t,Yk,t) Can be calculated by an automobile dynamic model, and the position constraint is determined as
Figure BDA0001459254100000032
Figure BDA0001459254100000033
Figure BDA0001459254100000034
Figure BDA0001459254100000035
Wherein a is the distance from the mass center of the automobile to the automobile head; b is the distance from the mass center of the automobile to the tail of the automobile; c is half of the width of the automobile;
Figure BDA0001459254100000036
predicting the yaw angle of the automobile at the k moment in the time domain by taking the t moment as a starting point; dx,j,tThe longitudinal distance from the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system, Dy,j,tThe transverse distance from the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system;
step 1.3, constructing a path dynamic programming multi-objective optimization control problem, solving the multi-objective optimization control problem, and further solving a yaw angular velocity reference value, a lateral displacement reference value and a longitudinal velocity reference value, wherein the method comprises the following substeps:
step 1.3.1, obtaining obstacle information through a radar sensor, obtaining automobile running state information through a vehicle speed sensor and a gyroscope, and inputting the obtained obstacle information and the automobile running state information into a path dynamic planning module;
step 1.3.2, converting the tracking performance index and the automobile safety index into a single index by using a linear weighting method, constructing a path dynamic planning multi-target optimization control problem which simultaneously meets automobile stability constraint and position constraint and ensures that path dynamic planning input and output conform to a particle model:
Figure BDA0001459254100000041
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
Step 1.3.3, in the path dynamic programming controller, calling a genetic algorithm, solving a multi-objective optimization control problem (8) to obtain the optimal open-loop control ay *Comprises the following steps:
Figure BDA0001459254100000042
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
Step 1.3.4, utilizing the optimal open loop control a at the current momenty *(0) To find a yaw angular velocity reference value
Figure BDA00014592541000000416
Reference value of yaw angle
Figure BDA0001459254100000043
Reference value of lateral displacement YrefLongitudinal speed reference value
Figure BDA0001459254100000044
The specific expression is as follows:
Figure BDA0001459254100000045
Figure BDA0001459254100000046
Figure BDA0001459254100000047
Figure BDA0001459254100000048
Figure BDA0001459254100000049
Figure BDA00014592541000000410
wherein V is the current longitudinal speed of the automobile,
Figure BDA00014592541000000411
is a reference value of the lateral speed of the automobile,
Figure BDA00014592541000000412
a reference value of the lateral speed of the path;
step 2, the path tracking control module receives a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of the expected track transmitted by the path dynamic planning module, and simultaneously, the path tracking control module collects current automobile driving state information, optimizes the front wheel turning angle and the four wheel slip rates of the automobile in real time, and controls the automobile to realize collision avoidance, and the path tracking control module comprises the following substeps:
step 2.1, the performance index design process of path tracking control comprises the following substeps:
step 2.1.1, utilizing the lateral displacement reference value Y output by the path dynamic programming modulerefYaw angle reference value
Figure BDA00014592541000000413
Yaw rate reference value
Figure BDA00014592541000000414
Longitudinal speed reference value
Figure BDA00014592541000000415
The two norms of the error of the actual automobile running state information are used as tracking performance indexes to reflect the tracking characteristics of the automobile, and the expression is as follows:
Figure BDA0001459254100000051
wherein, ηk，tIs the information of the running state of the automobile,
Figure BDA0001459254100000052
ηrefk，tthe reference value provided for the path dynamic programming module,
Figure BDA0001459254100000053
Hp，lfor predicting time domain of path tracking control module, w2Is a weight coefficient;
the automobile dynamic model comprises the following steps:
Figure BDA0001459254100000054
Figure BDA0001459254100000055
Figure BDA0001459254100000056
Fxi＝fxicos(i)-fyisin(i), i∈{1,2,3,4} (30)
Fyi＝fxisin(i)+fyicos(i), i∈{1,2,3,4} (31)
Figure BDA0001459254100000057
Figure BDA0001459254100000058
wherein, Fxi、FyiLongitudinal component force and lateral component force of the four wheels along the coordinate direction of the vehicle body are respectively obtained; f. ofxi、fyiComponent forces of the four wheels in the wheel coordinate direction, respectively, where fxiAs a function of the four wheel slip rates and wheel vertical loads, fyiThe specific value can be determined by a magic formula as a function of the front wheel rotation angle and the wheel vertical load;
Figure BDA0001459254100000059
respectively the longitudinal speed and the longitudinal acceleration of the automobile;
Figure BDA00014592541000000510
the lateral speed and the lateral acceleration of the automobile are respectively;
Figure BDA00014592541000000511
respectively representing the automobile yaw angle, the yaw angular velocity and the yaw angular acceleration; lf、lrRespectively the distances from the mass center of the automobile to the front, back and axis, lsHalf of the track width; j. the design is a squarezIs the yaw moment of inertia around the vertical axis of the center of mass of the automobile; m is the mass of the automobile; x, Y are respectively the horizontal and vertical coordinates of the position of the center of mass of the automobile in the geodetic coordinate system;iat four wheel corners, where the vehicle is front-wheel steered, so3＝4＝0；
The parameters of the magic formula are obtained by experimental fitting, and the specific expression is as follows:
Figure BDA00014592541000000512
Figure BDA00014592541000000513
Figure BDA00014592541000000514
Figure BDA00014592541000000515
Figure BDA00014592541000000516
Figure BDA0001459254100000061
Figure BDA0001459254100000062
Figure BDA0001459254100000063
Figure BDA0001459254100000064
Figure BDA0001459254100000065
Figure BDA0001459254100000066
Figure BDA0001459254100000067
wherein V is the current longitudinal speed of the automobile αf、αrRespectively a front wheel side deflection angle and a rear wheel side deflection angle; fz,f、Fz,rRespectively the front and rear axle loads of the automobile; siFor slipping of four wheels of a vehicleRate; a. thexi、Bxi、Cxi、Dxi、ExiAnd Ayi、Byi、Cyi、Dyi、EyiAre test fitting parameters, the specific parameters are shown in the following table:
TABLE 4 magic formula parameters
a0 a1 a2 a3 a4 a5 a6
1.75 0 1000 1289 7.11 0.0053 0.1925
b0 b1 b2 b3 b4 b5 b6 b7 b8
1.57 35 1200 60 300 0.17 0 0 0.2
Step 2.1.2, using the two-norm of the control quantity change rate as a steering brake smooth index of an actuator in the collision avoidance process to embody the steering brake smooth characteristic; the control quantity u is the rotation angle of the front wheel of the automobile and the slip rate s of the four wheels of the automobileii ∈ {1,2,3,4}, establishing a discrete quadratic steering brake smoothness index as:
Figure BDA0001459254100000071
wherein Hc，lFor controlling the time domain, t represents the current moment, and delta u is the change rate of the controlled variable;
step 2.2, designing the constraint of path tracking control as setting automobile stability constraint to ensure the safety of automobile obstacle avoidance; the method comprises the following steps of utilizing a linear inequality to limit the corner of a front wheel and the upper limit and the lower limit of the slip rates of four wheels to obtain the physical constraints of a steering actuator and a braking actuator, wherein the mathematical expression is as follows:
min＜k，t＜maxk＝t，t+1……t+Hc，l-11(23)
simin＜sik，t＜simaxi∈{1，2，3，4} k＝t，t+1……t+Hc，l-1 (24)
wherein the content of the first and second substances,minis the lower limit of the front wheel steering angle,maxis the upper limit of the front wheel steering angle, siminLower limit of slip ratio of four wheels, simaxThe upper limit of the slip rate of four wheels;
step 2.3, constructing a path tracking control multi-objective optimization control problem, solving the multi-objective optimization control problem, obtaining the corner of the front wheel of the automobile and the slip rates of four wheels of the automobile which are optimized in real time, and realizing the emergency collision avoidance control of the automobile, wherein the method comprises the following substeps:
step 2.3.1, the path tracking control module obtains a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of the expected track from the path dynamic planning module;
step 2.3.2, converting the tracking performance index and the steering brake smooth index into a single index by using a linear weighting method, and constructing a path tracking control multi-target optimization control problem which needs to meet the physical constraints of a steering actuator and a brake actuator at the same time and ensure that the path tracking control input and output conform to an automobile dynamics model:
Figure BDA0001459254100000072
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
Step 2.3.3, in the path tracking controller, calling a genetic algorithm, solving a multi-objective optimization control problem (25), and obtaining the optimal open-loop control u as:
Figure BDA0001459254100000073
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
Step 2.3.4, utilizing the optimal open loop control u at the current moment*(0) Feedback is carried out to realize closed-loop control, the corner of the front wheel of the automobile and the slip ratio of four wheels of the automobile which are optimized in real time are obtained, and the emergency collision avoidance control of the automobile is realized;
step 3, designing an EPS moment compensation module implanted with a steering wheel sudden change moment humanized adjustment algorithm, determining a compensation control moment by the EPS moment compensation module according to the vehicle speed and the additional rotation angle of the front wheel, and controlling the steering wheel sudden change moment within the acceptable range of a driver; the front wheel additional corner is a difference value between a front wheel corner optimized by the path tracking control module and a front wheel corner generated by steering input of a driver, and is realized by an AFS (automatic navigation System); the design process includes the following substeps:
step 3.1, the design method of the EPS moment compensation module comprises the following steps: selecting a plurality of drivers to carry out real-vehicle debugging, and firstly, debugging the speed and the moment compensation control gain under the additional turning angle of the front wheel by debugging, and repeatedly debugging the drivers according to the subjective feeling of the drivers by the experimenter to ensure that the sudden change moment of the steering wheel can be accepted by the drivers;
3.2, changing the additional turning angles of the front wheels, debugging the moment compensation control gain by an experimenter to enable the steering wheel sudden change moment under the intervention of the additional turning angles of the different front wheels to be accepted by a driver, and further determining the moment compensation control gain under the vehicle speed;
step 3.3, determining torque compensation control gains under the intervention of different vehicle speeds and different front wheel additional rotation angles by adopting the same method, and completing the determination of the three-dimensional MAP of the vehicle speed, the front wheel additional rotation angles and the torque compensation control gains;
and 3.4, implanting the EPS moment compensation control gain three-dimensional MAP into an EPS controller, and controlling an EPS power-assisted motor by the EPS controller to achieve the control effect of moment compensation.
The invention has the beneficial effects that: by constructing a layered optimization problem based on model prediction control, the upper layer adopts a particle model to carry out path planning, and the lower layer adopts a high-precision automobile dynamics model to carry out path tracking, the problems of path dynamic planning and real-time tracking during emergency collision avoidance are solved, and safe optimal collision avoidance is realized. And the steering wheel sudden change torque is controlled within an ideal range through the EPS torque compensation controller. According to the method, the upper-layer path dynamic planning module takes the shortest collision avoidance distance as an optimization target and takes no collision as a constraint condition, so that the real-time performance of path dynamic planning can be effectively improved. Meanwhile, the method repeatedly debugs the EPS moment compensation control gain by using a subjective evaluation mode, thereby realizing humanized abrupt moment adjustment.
Drawings
FIG. 1 is a schematic diagram of the present invention of a layered control method for emergency collision avoidance of an automobile considering man-machine harmony.
Fig. 2 is a schematic diagram of the relationship between the position of the vehicle and the position of the obstacle.
FIG. 3 is a diagram of an automobile model according to the present invention.
FIG. 4 is a schematic diagram of an EPS moment compensation controller experiment process of the present invention.
FIG. 5 is a three-dimensional MAP graph of EPS torque compensation control gain of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
As shown in FIG. 1, the automobile emergency collision avoidance hierarchical control method considering man-machine harmony of the invention comprises the following steps: the path dynamic planning module 1 obtains a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track through real-time optimization according to barrier information, target point coordinates and automobile running state information which are collected in real time, inputs the lateral displacement reference value, the yaw angle speed reference value and the longitudinal speed reference value into the path tracking control module 2, meanwhile, the path tracking control module 2 collects current automobile running state information, obtains a front wheel turning angle and four wheel slip rates of an automobile 3 through real-time optimization, and controls the automobile 3 to assist a driver 5 in collision avoidance; in the process of controlling collision avoidance, the EPS moment compensation module 4 determines a compensation control moment according to the vehicle speed and the additional rotation angle of the front wheel, and controls the sudden change moment of the steering wheel within an ideal range to realize the man-machine harmonious emergency collision avoidance of the automobile. The obstacle information comprises discrete point coordinates of the outline of the obstacle, and is obtained by measuring by a radar sensor; the automobile running state information comprises the automobile longitudinal speed, the lateral speed and the yaw rate, wherein the automobile longitudinal speed and the lateral speed are measured by the automobile speed sensor, and the automobile yaw rate is measured by the gyroscope.
The method of the present invention is specifically described below with a car as a platform, and the main parameters of the test car are shown in table 1:
table 1 main parameters of the test car
Figure BDA0001459254100000091
The path dynamic planning module 1 realizes the following three functions: 1.1, designing a performance index of dynamic path planning; 1.2, designing the constraint of dynamic path planning; and 1.3, performing rolling time domain solution on a path dynamic programming control law.
In section 1.1, the performance index design of the dynamic path planning includes the following two contents: 1.1.1, utilizing a two-norm of the error between the terminal point coordinate of the predicted track in the predicted time domain and the coordinate of the target point as a tracking performance index to embody the track tracking characteristic of the automobile; 1.1.2, utilizing the two-norm of the lateral acceleration as an automobile safety index to embody the collision avoidance stability of the automobile;
in section 1.1.1, tracking the performance index by taking the two-norm of the error between the endpoint coordinate of the predicted track in the predicted time domain and the coordinate of the target point as an evaluation standard, wherein the expression is as follows:
Figure BDA0001459254100000101
wherein Hp，hFor the prediction time domain of the dynamic path planning module, (X)t+Hp，h,Yt+Hp，h) To predictThe terminal point coordinate of the predicted track in the time domain is obtained by the iteration of the automobile dynamic model, and the coordinate (X) of the target point to be reached by the automobile in collision avoidanceg,Yg) I.e. a safety point behind an obstacle.
In the 1.1.2 part, the automobile collision avoidance stability in the collision avoidance process is described by utilizing the two norms of the lateral acceleration, and the discrete quadratic automobile safety index is established as follows:
Figure BDA0001459254100000102
wherein Hc，hA control time domain of a dynamic path planning module, t represents the current moment, ayIs the lateral acceleration, w, of the particle model1Is ayThe weight coefficients and the design parameters of the dynamic path planning module are shown in Table 2, wherein Ts1The sampling period of the module is dynamically planned for the path.
TABLE 2 Emergency Collision avoidance controller design parameters
Controller parameters Parameter value Controller parameters Parameter value
Hp，h 5 Hc，h 2
w1 0.5 Ts1 0.01s
In section 1.2, the constrained design of path dynamic planning includes two parts: 1.2.1, setting automobile stability constraint to ensure the safety of automobile obstacle avoidance; 1.2.2, position restraint is arranged, and collision with an obstacle is avoided in the collision avoidance process.
In section 1.2.1, the stability constraint of the automobile is obtained by limiting the upper limit and the lower limit of the lateral acceleration by using a linear inequality, and the mathematical expression is as follows:
|ayk，t|＜μg k＝t，t+1……t+Hc，h-1 (3)
where μ is a road surface adhesion coefficient obtained by the estimator, and g is a gravitational acceleration.
At section 1.2.2, as shown in FIG. 2, the position information of the obstacle at time t may be characterized as a set of N discrete points, which may be measured by the radar sensor, where the coordinate of the jth discrete point is expressed as (X)j,t,Yj,t) And the coordinate of the mass center of the automobile at the moment t is recorded as (X)k,t,Yk,t) Can be calculated by an automobile dynamic model, and the position constraint is determined as
Figure BDA0001459254100000103
Figure BDA0001459254100000104
Figure BDA0001459254100000105
Figure BDA0001459254100000106
Wherein a is the distance from the mass center of the automobile to the automobile head; b is the distance from the mass center of the automobile to the tail of the automobile; c is half of the width of the automobile;
Figure BDA0001459254100000111
for predicting the yaw angle of the vehicle at the time k in the time domain starting at the time t, Dx,j,tThe longitudinal distance from the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system, Dy,j,tThe lateral distance of the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system.
In section 1.3, the rolling time domain solution of the path dynamic programming control law comprises the following steps:
1.3.1, acquiring obstacle information through a radar sensor, acquiring automobile running state information through a vehicle speed sensor and a gyroscope, and inputting the acquired obstacle information and the automobile running state information into a path dynamic planning module 1;
1.3.2, converting the tracking performance index and the automobile safety index into a single index by using a linear weighting method, constructing a path dynamic planning multi-target optimization control problem, wherein the problem simultaneously meets automobile stability constraint and position constraint, and ensures that path dynamic planning input and output conform to a particle model:
Figure BDA0001459254100000112
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
1.3.3, in the path dynamic programming controller, calling a genetic algorithm, solving a multi-objective optimization control problem (8) to obtain the optimal open-loop control ay *Comprises the following steps:
Figure BDA0001459254100000113
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
1.3.4, utilizing the optimal open loop control a at the current momenty *(0) To find a yaw angular velocity reference value
Figure BDA0001459254100000114
Reference value of yaw angle
Figure BDA0001459254100000115
Reference value of lateral displacement YrefLongitudinal speed reference value
Figure BDA0001459254100000116
The specific expression is as follows:
Figure BDA0001459254100000117
Figure BDA0001459254100000118
Figure BDA0001459254100000119
Figure BDA00014592541000001110
Figure BDA00014592541000001111
Figure BDA00014592541000001112
wherein V is the current longitudinal speed of the automobile,
Figure BDA00014592541000001113
is a reference value of the lateral speed of the automobile,
Figure BDA00014592541000001114
is a reference value for the lateral velocity of the path.
The particle model is:
Figure BDA0001459254100000121
Figure BDA0001459254100000122
Figure BDA0001459254100000123
Figure BDA0001459254100000124
Figure BDA0001459254100000125
wherein the content of the first and second substances,
Figure BDA0001459254100000126
ayis the lateral acceleration speed of the automobile;
Figure BDA0001459254100000127
is the longitudinal acceleration of the vehicle;
Figure BDA0001459254100000128
the yaw angle and the yaw angular velocity of the automobile;
Figure BDA0001459254100000129
respectively the longitudinal speed and the lateral speed of the mass center of the automobile in a geodetic coordinate system.
The path tracking control module 2 realizes the following three functions: 2.1, designing a performance index of path tracking control; 2.2, constrained design of path tracking control; and 2.3, performing rolling time domain solution on the path tracking control law.
In section 2.1, the performance index design of path tracking control includes the following two contents: 2.1.1, utilizing the lateral displacement reference value Y output by the path dynamic programming modulerefYaw angle reference value
Figure BDA00014592541000001210
Yaw rate reference value
Figure BDA00014592541000001211
Longitudinal speed reference value
Figure BDA00014592541000001212
The two norms of the error of the actual automobile running state information are used as tracking performance indexes to reflect the tracking characteristics of the automobile; and 2.1.2, utilizing the two-norm of the control quantity change rate as a steering brake smooth index to embody the steering brake smooth characteristic.
In section 2.1.1, the tracking performance index takes a two-norm of an error between a reference value output by the path dynamic planning module and actual automobile running state information as an evaluation standard, and an expression is as follows:
Figure BDA00014592541000001213
wherein, ηk，tIs the information of the running state of the automobile,
Figure BDA00014592541000001214
ηrefk，tthe reference value provided for the path dynamic programming module,
Figure BDA00014592541000001215
Hp，lfor predicting time domain of path tracking control module, w2Are weight coefficients.
In section 2.1.2, the smooth characteristic of steering brake of the actuator in the collision avoidance process is described by utilizing a two-norm of the change rate of the control quantity u, wherein the control quantity u is the rotation angle of the front wheel of the automobile and the slip rate s of the four wheels of the automobileii ∈ {1,2,3,4}, establishing a discrete quadratic steering brake smoothness index as:
Figure BDA00014592541000001216
wherein Hc,lFor control of the time domain, t denotes the current time, Δ u is controlThe variable rate, path-tracking control module design parameters are shown in Table 3, where Ts2The sampling period of the control module is tracked for the path.
TABLE 3 Emergency Collision avoidance controller design parameters
Figure BDA00014592541000001217
Figure BDA0001459254100000131
In the 2.2 part, the constraint of the dynamic path planning is designed to set the stability constraint of the automobile, so that the safety of the automobile in obstacle avoidance is guaranteed; the method comprises the following steps of utilizing a linear inequality to limit the corner of a front wheel and the upper limit and the lower limit of the slip rates of four wheels to obtain the physical constraints of a steering actuator and a braking actuator, wherein the mathematical expression is as follows:
min＜k，t＜maxk＝t，t+1……t+Hc，l-1 (23)
simin＜sik，t＜simaxi∈{1，2，3，4} k＝t，t+1……t+Hc，l-1 (24)
wherein the content of the first and second substances,minis the lower limit of the front wheel steering angle,maxis the upper limit of the front wheel steering angle, siminLower limit of slip ratio of four wheels, simaxThe upper limit of the slip ratio of four wheels.
In section 2.3, the rolling time domain solution of the path tracking control law comprises the following steps:
2.3.1, obtaining a reference value from the path dynamic planning module, and inputting information into the path tracking control module;
2.3.2, converting the tracking performance index and the steering braking smooth index into a single index by using a linear weighting method, and constructing a path tracking control multi-target optimization control problem which needs to meet the physical constraints of steering and braking actuators at the same time and ensure that the path tracking control input and output conform to an automobile dynamics model:
Figure BDA0001459254100000132
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
2.3.3, in the path tracking controller, calling a genetic algorithm to solve a multi-objective optimization control problem (25) to obtain the optimal open-loop control u*Comprises the following steps:
Figure BDA0001459254100000133
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
2.3.4, utilizing the optimal open loop control u at the current moment*(0) Feedback is carried out to realize closed-loop control;
as shown in fig. 3, the dynamic model of the automobile according to the present invention is:
Figure BDA0001459254100000134
Figure BDA0001459254100000135
Figure BDA0001459254100000141
Fxi＝fxicos(i)-fyisin(i), i∈{1,2,3,4} (30)
Fyi＝fxisin(i)+fyicos(i), i∈{1,2,3,4} (31)
Figure BDA0001459254100000142
Figure BDA0001459254100000143
wherein, Fxi、FyiLongitudinal component force and lateral component force of the four wheels along the coordinate direction of the vehicle body are respectively obtained; f. ofxi、fyiComponent forces of the four wheels in the wheel coordinate direction, respectively, where fxiAs a function of the four wheel slip rates and wheel vertical loads, fyiThe specific value can be determined by a magic formula as a function of the front wheel rotation angle and the wheel vertical load;
Figure BDA0001459254100000144
respectively the longitudinal speed and the longitudinal acceleration of the automobile;
Figure BDA0001459254100000145
the lateral speed and the lateral acceleration of the automobile are respectively;
Figure BDA0001459254100000146
respectively representing the automobile yaw angle, the yaw angular velocity and the yaw angular acceleration; lf、lrRespectively the distance from the center of mass of the automobile to the front and rear axles lsHalf of the track width; j. the design is a squarezIs the yaw moment of inertia around the vertical axis of the center of mass of the automobile; m is the mass of the automobile; x, Y are respectively the horizontal and vertical coordinates of the position of the center of mass of the automobile in the geodetic coordinate system;iat four wheel corners, where the vehicle is front-wheel steered, so3＝4＝0；
The parameters of the magic formula are obtained by experimental fitting, and the specific expression is as follows:
Figure BDA0001459254100000147
Figure BDA0001459254100000148
Figure BDA0001459254100000149
Figure BDA00014592541000001410
Figure BDA00014592541000001411
Figure BDA00014592541000001412
Figure BDA00014592541000001413
Figure BDA0001459254100000151
Figure BDA0001459254100000152
Figure BDA0001459254100000153
Figure BDA0001459254100000154
Figure BDA0001459254100000155
wherein V is the current longitudinal speed of the automobile αf、αrRespectively a front wheel side deflection angle and a rear wheel side deflection angle; fz,f、Fz,rRespectively the front and rear axle loads of the automobile; siThe slip rate of four wheels of the automobile; a. thexi、Bxi、Cxi、Dxi、ExiAnd Ayi、Byi、Cyi、Dyi、EyiAre test fitting parameters, the specific parameters are shown in the following table:
TABLE 4 parameter value-taking table of magic formula
a0 a1 a2 a3 a4 a5 a6
1.75 0 1000 1289 7.11 0.0053 0.1925
b0 b1 b2 b3 b4 b5 b6 b7 b8
1.57 35 1200 60 300 0.17 0 0 0.2
The design method of the EPS moment compensation module 4 comprises the following steps: 30 drivers are selected and classified into the following four categories according to gender and proficiency: a skilled male driver, a skilled female driver, an unskilled male driver, and an unskilled female driver. The driver respectively carries out real vehicle debugging according to the pre-classification, the debugging process is shown in figure 4, firstly, the vehicle speed is set to be 60km/h, the additional turning angle of a front wheel is set to be 3deg, an experimenter repeatedly debugs the sudden change torque compensation control gain according to feedback information of the acceptance degree of the driver to the sudden change torque of the steering wheel, when the driver feels that the sudden change torque is overlarge, the experimenter reduces the torque compensation control gain, when the driver feels that the sudden change torque is overlarge, the experimenter adjusts the torque compensation control gain to be large, finally, the situation that the sudden change torque of the steering wheel can be accepted by the driver is ensured, and the torque compensation control gain value at the moment is recorded; secondly, the speed is still determined to be 60km/h, the additional turning angle range of the front wheels is-6 deg to 6deg, the interval is 2deg, the left side and the right side are symmetrical when the automobile steers, and the abrupt change moments of the steering wheel generated on the left side and the right side are the same under the condition that the additional turning angles of the front wheels have the same amplitude, so that the moment compensation control gain can be obtained only by adjusting the additional turning angle range of the front wheels to be 0deg to 6 deg. During testing, an experimenter debugs moment compensation control gains under the intervention of each corner in a range of 0deg to 6deg according to the acceptance degree of the driver to the steering wheel sudden change moment, so that the steering wheel sudden change moment under the intervention of the additional corner of each front wheel is accepted by the driver, the moment compensation control gains under the intervention of different corners at the speed of 60km/h are further determined, and specific numerical values of the moment compensation control gains are recorded; finally, torque compensation control gains under the intervention of different turning angles at different vehicle speeds are debugged by the same method, the vehicle speed range is 10km/h to 100km/h, the vehicle speed interval is 20km/h, and finally a three-dimensional numerical table of the vehicle speed, the additional turning angle of the front wheel and the torque compensation control gain is determined, and fig. 5 is an EPS torque compensation control gain three-dimensional MAP graph. And finally, implanting the EPS moment compensation control gain three-dimensional MAP into an EPS controller, and controlling an EPS power-assisted motor by the EPS controller to achieve the control effect of moment compensation.

Claims (1)

1. An automobile emergency collision avoidance layered control method considering man-machine harmony is characterized in that the method comprises the following steps: the method comprises the steps that a path dynamic planning module optimizes in real time to obtain a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track according to barrier information, target point coordinates and automobile running state information which are collected in real time, the lateral displacement reference value, the yaw angle speed reference value and the longitudinal speed reference value are input into a path tracking control module, meanwhile, the path tracking control module collects current automobile running state information, front wheel corners and automobile four-wheel slip rates are obtained through real-time optimization, and the automobile is controlled to avoid collision; in the collision avoidance control process, the EPS moment compensation module determines compensation control moment according to the vehicle speed and the additional rotation angle of the front wheel, and controls the sudden change moment of the steering wheel within an ideal range to realize the emergency collision avoidance of the man-machine harmonious automobile; the method comprises the following steps:
step 1, a path dynamic planning module optimizes in real time to obtain a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of an expected track according to barrier information, target point coordinates and automobile running state information which are collected in real time, and the path dynamic planning module comprises the following substeps:
step 1.1, the performance index design process of the dynamic path planning comprises the following substeps:
step 1.1.1, using a two-norm of the error between the terminal point coordinate of the predicted track in the predicted time domain and the target point coordinate as a tracking performance index to reflect the track tracking characteristic of the automobile, wherein the expression is as follows:
Figure FDA0001459254090000011
wherein Hp，hFor the prediction time domain of the dynamic path planning module, (X)t+Hp，h,Yt+Hp，h) The coordinates (X) of the target point to be reached by the automobile in collision avoidance are obtained by iteration of the particle model for predicting the terminal point coordinates of the predicted track in the time domaing,Yg)；
The particle model is:
Figure FDA0001459254090000012
Figure FDA0001459254090000013
Figure FDA0001459254090000014
Figure FDA0001459254090000015
Figure FDA0001459254090000016
wherein the content of the first and second substances,
Figure FDA0001459254090000017
ayis the lateral acceleration speed of the automobile;
Figure FDA0001459254090000018
is the longitudinal acceleration of the vehicle;
Figure FDA0001459254090000019
respectively the automobile yaw angle and the yaw angular speed;
Figure FDA00014592540900000110
respectively the longitudinal speed and the lateral speed of the mass center of the automobile in a geodetic coordinate system; v is the current longitudinal speed of the automobile;
step 1.1.2, utilizing the two-norm of the lateral acceleration as the automobile safety index in the collision avoidance process to embody the automobile collision avoidance stability, and establishing the discrete quadratic automobile safety index as follows:
Figure FDA00014592540900000111
wherein Hc，hA control time domain of a dynamic path planning module, t represents the current moment, ayIs the lateral acceleration, w, of the particle model1Is ayThe weight coefficient of (a);
step 1.2, the constraint design process of the dynamic path planning comprises the following substeps:
step 1.2.1, setting automobile stability constraint to ensure the safety of automobile obstacle avoidance;
and (3) limiting the upper limit and the lower limit of the lateral acceleration by using a linear inequality to obtain the stability constraint of the automobile, wherein the mathematical expression is as follows:
|ayk，t|＜μg k＝t，t+1……t+Hc，h-1 (3)
wherein mu is a road surface adhesion coefficient, and g is a gravity acceleration;
step 1.2.2, setting position constraint to ensure that the collision with an obstacle is avoided in the collision avoidance process;
the position information of the obstacle at time t may be characterized as a set of N discrete points, which may be determined by a radar sensorMeasured, where the coordinates of the jth discrete point are expressed as (X)j,t,Yj,t) And the coordinate of the mass center of the automobile at the moment t is recorded as (X)k,t,Yk,t) Can be calculated by an automobile dynamic model, and the position constraint is determined as
Figure FDA0001459254090000021
Figure FDA0001459254090000022
Figure FDA0001459254090000023
Figure FDA0001459254090000024
Wherein a is the distance from the mass center of the automobile to the automobile head; b is the distance from the mass center of the automobile to the tail of the automobile; c is half of the width of the automobile;
Figure FDA0001459254090000026
predicting the yaw angle of the automobile at the k moment in the time domain by taking the t moment as a starting point; dx,j,tThe longitudinal distance from the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system, Dy,j,tThe transverse distance from the jth discrete point of the obstacle to the center of mass of the automobile in the automobile coordinate system;
step 1.3, constructing a path dynamic programming multi-objective optimization control problem, solving the multi-objective optimization control problem, and further solving a yaw angular velocity reference value, a lateral displacement reference value and a longitudinal velocity reference value, wherein the method comprises the following substeps:
step 1.3.1, obtaining obstacle information through a radar sensor, obtaining automobile running state information through a vehicle speed sensor and a gyroscope, and inputting the obtained obstacle information and the automobile running state information into a path dynamic planning module;
step 1.3.2, converting the tracking performance index and the automobile safety index into a single index by using a linear weighting method, constructing a path dynamic planning multi-target optimization control problem which simultaneously meets automobile stability constraint and position constraint and ensures that path dynamic planning input and output conform to a particle model:
Figure FDA0001459254090000025
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
Step 1.3.3, in the path dynamic programming controller, calling a genetic algorithm, solving a multi-objective optimization control problem (8) to obtain the optimal open-loop control ay *Comprises the following steps:
Figure FDA0001459254090000031
subject to
i) Particle model
ii) the constraint conditions are equations (3) to (7)
Step 1.3.4, utilizing the optimal open loop control a at the current momenty *(0) To find a yaw angular velocity reference value
Figure FDA0001459254090000032
Reference value of yaw angle
Figure FDA0001459254090000033
Reference value of lateral displacement YrefLongitudinal speed reference value
Figure FDA0001459254090000034
The specific expression is as follows:
Figure FDA0001459254090000035
Figure FDA00014592540900000319
Figure FDA0001459254090000036
Figure FDA0001459254090000037
Figure FDA0001459254090000038
Figure FDA0001459254090000039
wherein V is the current longitudinal speed of the automobile,
Figure FDA00014592540900000310
is a reference value of the lateral speed of the automobile,
Figure FDA00014592540900000311
a reference value of the lateral speed of the path;
step 2, the path tracking control module receives a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of the expected track transmitted by the path dynamic planning module, and simultaneously, the path tracking control module collects current automobile driving state information, optimizes the front wheel turning angle and the four wheel slip rates of the automobile in real time, and controls the automobile to realize collision avoidance, and the path tracking control module comprises the following substeps:
step 2.1, the performance index design process of path tracking control comprises the following substeps:
step 2.1.1, utilizing the lateral displacement reference value Y output by the path dynamic programming modulerefYaw angle reference value
Figure FDA00014592540900000312
Yaw rate reference value
Figure FDA00014592540900000313
Longitudinal speed reference value
Figure FDA00014592540900000314
The two norms of the error of the actual automobile running state information are used as tracking performance indexes to reflect the tracking characteristics of the automobile, and the expression is as follows:
Figure FDA00014592540900000315
wherein, ηk，tIs the information of the running state of the automobile,
Figure FDA00014592540900000316
ηrefk，tthe reference value provided for the path dynamic programming module,
Figure FDA00014592540900000317
Hp，lfor predicting time domain of path tracking control module, w2Is a weight coefficient;
the automobile dynamic model comprises the following steps:
Figure FDA00014592540900000318
Figure FDA0001459254090000041
Figure FDA0001459254090000042
Fxi＝fxicos(i)-fyisin(i),i∈{1,2,3,4} (30)
Fyi＝fxisin(i)+fyicos(i),i∈{1,2,3,4} (31)
Figure FDA0001459254090000043
Figure FDA0001459254090000044
wherein, Fxi、FyiLongitudinal component force and lateral component force of the four wheels along the coordinate direction of the vehicle body are respectively obtained; f. ofxi、fyiComponent forces of the four wheels in the wheel coordinate direction, respectively, where fxiAs a function of the four wheel slip rates and wheel vertical loads, fyiThe specific value can be determined by a magic formula as a function of the front wheel rotation angle and the wheel vertical load;
Figure FDA0001459254090000045
respectively the longitudinal speed and the longitudinal acceleration of the automobile;
Figure FDA0001459254090000046
the lateral speed and the lateral acceleration of the automobile are respectively;
Figure FDA0001459254090000047
respectively representing the automobile yaw angle, the yaw angular velocity and the yaw angular acceleration; lf、lrRespectively the distances from the mass center of the automobile to the front, back and axis, lsHalf of the track width; j. the design is a squarezIs the yaw moment of inertia around the vertical axis of the center of mass of the automobile; m is the mass of the automobile; x, Y are respectively the horizontal and vertical coordinates of the position of the center of mass of the automobile in the geodetic coordinate system;iat four wheel corners, where the vehicle is front-wheel steered, so3＝4＝0；
The parameters of the magic formula are obtained by experimental fitting, and the specific expression is as follows:
Figure FDA0001459254090000048
Figure FDA0001459254090000049
Figure FDA00014592540900000410
Figure FDA00014592540900000411
Figure FDA00014592540900000412
Figure FDA00014592540900000413
Figure FDA00014592540900000414
Figure FDA0001459254090000051
Figure FDA0001459254090000052
Figure FDA0001459254090000053
Figure FDA0001459254090000054
Figure FDA0001459254090000055
wherein V is the current longitudinal speed of the automobile αf、αrRespectively a front wheel side deflection angle and a rear wheel side deflection angle; fz,f、Fz,rRespectively the front and rear axle loads of the automobile; siThe slip rate of four wheels of the automobile; a. thexi、Bxi、Cxi、Dxi、ExiAnd Ayi、Byi、Cyi、Dyi、EyiAre test fitting parameters, the specific parameters are shown in the following table:
TABLE 4 magic formula parameters
a0 a1 a2 a3 a4 a5 a6 1.75 0 1000 1289 7.11 0.0053 0.1925 b0 b1 b2 b3 b4 b5 b6 b7 b8 1.57 35 1200 60 300 0.17 0 0 0.2
Step 2.1.2, using the two-norm of the control quantity change rate as a steering brake smooth index of an actuator in the collision avoidance process to embody the steering brake smooth characteristic; the control quantity u is the rotation angle of the front wheel of the automobile and the slip rate s of the four wheels of the automobileii ∈ {1,2,3,4}, establishing a discrete quadratic steering systemThe dynamic smoothing index is:
Figure FDA0001459254090000056
wherein Hc，lFor controlling the time domain, t represents the current moment, and delta u is the change rate of the controlled variable;
step 2.2, designing the constraint of path tracking control as setting automobile stability constraint to ensure the safety of automobile obstacle avoidance; the method comprises the following steps of utilizing a linear inequality to limit the corner of a front wheel and the upper limit and the lower limit of the slip rates of four wheels to obtain the physical constraints of a steering actuator and a braking actuator, wherein the mathematical expression is as follows:
min＜k，t＜maxk＝t，t+1……t+Hc，l-1 (23)
simin＜sik，t＜simaxi∈{1，2，34} k＝t，t+1……t+Hc，l-1 (24)
wherein the content of the first and second substances,minis the lower limit of the front wheel steering angle,maxis the upper limit of the front wheel steering angle, siminLower limit of slip ratio of four wheels, simaxThe upper limit of the slip rate of four wheels;
step 2.3, constructing a path tracking control multi-objective optimization control problem, solving the multi-objective optimization control problem, obtaining the corner of the front wheel of the automobile and the slip rates of four wheels of the automobile which are optimized in real time, and realizing the emergency collision avoidance control of the automobile, wherein the method comprises the following substeps:
step 2.3.1, the path tracking control module obtains a lateral displacement reference value, a yaw angle speed reference value and a longitudinal speed reference value of the expected track from the path dynamic planning module;
step 2.3.2, converting the tracking performance index and the steering brake smooth index into a single index by using a linear weighting method, and constructing a path tracking control multi-target optimization control problem which needs to meet the physical constraints of a steering actuator and a brake actuator at the same time and ensure that the path tracking control input and output conform to an automobile dynamics model:
Figure FDA0001459254090000061
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
Step 2.3.3, in the path tracking controller, calling a genetic algorithm, solving a multi-objective optimization control problem (25) to obtain an optimal open-loop control u*Comprises the following steps:
Figure FDA0001459254090000062
subject to
i) Automobile dynamics model
ii) the constraint conditions are equations (23) to (24)
Step 2.3.4, utilizing the optimal open loop control u at the current moment*(0) Feedback is carried out to realize closed-loop control, the corner of the front wheel of the automobile and the slip ratio of four wheels of the automobile which are optimized in real time are obtained, and the emergency collision avoidance control of the automobile is realized;
step 3, designing an EPS moment compensation module implanted with a steering wheel sudden change moment humanized adjustment algorithm, determining a compensation control moment by the EPS moment compensation module according to the vehicle speed and the additional rotation angle of the front wheel, and controlling the steering wheel sudden change moment within the acceptable range of a driver; the front wheel additional corner is a difference value between a front wheel corner optimized by the path tracking control module and a front wheel corner generated by steering input of a driver, and is realized by an AFS (automatic navigation System); the design process includes the following substeps:
step 3.1, the design method of the EPS moment compensation module comprises the following steps: selecting a plurality of drivers to carry out real-vehicle debugging, and firstly, debugging the speed and the moment compensation control gain under the additional turning angle of the front wheel by debugging, and repeatedly debugging the drivers according to the subjective feeling of the drivers by the experimenter to ensure that the sudden change moment of the steering wheel can be accepted by the drivers;
3.2, changing the additional turning angles of the front wheels, debugging the moment compensation control gain by an experimenter to enable the steering wheel sudden change moment under the intervention of the additional turning angles of the different front wheels to be accepted by a driver, and further determining the moment compensation control gain under the vehicle speed;
step 3.3, determining torque compensation control gains under the intervention of different vehicle speeds and different front wheel additional rotation angles by adopting the same method, and completing the determination of the three-dimensional MAP of the vehicle speed, the front wheel additional rotation angles and the torque compensation control gains;
and 3.4, implanting the EPS moment compensation control gain three-dimensional MAP into an EPS controller, and controlling an EPS power-assisted motor by the EPS controller to achieve the control effect of moment compensation.
CN201711082166.2A 2017-11-07 2017-11-07 Automobile emergency collision avoidance layered control method considering man-machine harmony Expired - Fee Related CN107885932B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711082166.2A CN107885932B (en) 2017-11-07 2017-11-07 Automobile emergency collision avoidance layered control method considering man-machine harmony

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711082166.2A CN107885932B (en) 2017-11-07 2017-11-07 Automobile emergency collision avoidance layered control method considering man-machine harmony

Publications (2)

Publication Number Publication Date
CN107885932A CN107885932A (en) 2018-04-06
CN107885932B true CN107885932B (en) 2020-10-09

Family

ID=61779158

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711082166.2A Expired - Fee Related CN107885932B (en) 2017-11-07 2017-11-07 Automobile emergency collision avoidance layered control method considering man-machine harmony

Country Status (1)

Country Link
CN (1) CN107885932B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109002041B (en) * 2018-08-09 2021-03-19 北京智行者科技有限公司 Vehicle obstacle avoidance method
CN108860149B (en) * 2018-08-20 2020-02-07 中原工学院 Motion trajectory design method for shortest free lane change of intelligent vehicle
CN110626340B (en) * 2019-09-30 2023-07-11 南京航空航天大学 Intelligent automobile path tracking control system and method based on wolf algorithm
CN111391828B (en) * 2020-03-13 2021-06-15 南京航空航天大学 Method for planning vehicle collision avoidance track under pedestrian crossing working condition
CN111538328B (en) * 2020-04-03 2022-07-26 浙江工业大学 Priority hierarchical prediction control method for obstacle avoidance trajectory planning and tracking control of autonomous driving vehicle
CN111845738B (en) * 2020-06-22 2021-10-12 江苏大学 Vehicle path tracking control method based on double-model combination
CN111791898B (en) * 2020-08-13 2021-07-02 清华大学 Automatic driving automobile collision avoidance control method based on cooperation type game
CN112068445B (en) * 2020-09-23 2021-05-25 北京理工大学 Integrated control method and system for path planning and path tracking of automatic driving vehicle
CN112109705A (en) * 2020-09-23 2020-12-22 同济大学 Collision avoidance optimization control system and method for extended-range distributed driving electric vehicle
CN113104032B (en) * 2021-05-07 2022-04-12 大连理工大学 Active collision avoidance fault-tolerant system of distributed driving vehicle and working method thereof
CN113581201B (en) * 2021-07-22 2022-11-04 重庆邮电大学 Potential field model-based collision avoidance control method and system for unmanned vehicle
CN114407880B (en) * 2022-02-18 2023-06-27 岚图汽车科技有限公司 Unmanned emergency obstacle avoidance path tracking method

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105501216A (en) * 2016-01-25 2016-04-20 合肥工业大学 Internet of vehicles based hierarchical energy management control method for hybrid vehicle

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105501216A (en) * 2016-01-25 2016-04-20 合肥工业大学 Internet of vehicles based hierarchical energy management control method for hybrid vehicle

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Combined Speed and Steering Control in High-Speed Autonomous Ground Vehicles for Obstacle Avoidance Using Model Predictive Control;Jiechao Liu 等;《IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY》;20171031;第66卷(第10期);全文 *
Vehicle Dynamics Model & Path Stability Control Algorithms;Mathias Lidberg 等;《Interactive》;20131121;全文 *
汽车纵向避撞动力学模型的建立与仿真;孙海云 等;《汽车使用技术》;20130930;全文 *
电动汽车主动避撞控制系统建模与仿真研究;闫丹彤;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20170315;全文 *

Also Published As

Publication number Publication date
CN107885932A (en) 2018-04-06

Similar Documents

Publication Publication Date Title
CN107885932B (en) Automobile emergency collision avoidance layered control method considering man-machine harmony
JP6341137B2 (en) Vehicle driving support control device
JP6179820B2 (en) Vehicle driving support control device
Pérez et al. Cascade architecture for lateral control in autonomous vehicles
Attia et al. Coupled longitudinal and lateral control strategy improving lateral stability for autonomous vehicle
Hernandez et al. Steering control of automated vehicles using absolute positioning GPS and magnetic markers
Hima et al. Trajectory tracking for highly automated passenger vehicles
Cai et al. Implementation and development of a trajectory tracking control system for intelligent vehicle
Chen et al. Dynamic drifting control for general path tracking of autonomous vehicles
Vivek et al. A comparative study of Stanley, LQR and MPC controllers for path tracking application (ADAS/AD)
CN111923908A (en) Stability-fused intelligent automobile path tracking control method
CN111409641A (en) Trajectory tracking control method and system based on longitudinal and transverse coordination
CN110703754A (en) Path and speed highly-coupled trajectory planning method for automatic driving vehicle
CN111645755B (en) Control method and device
WO2022266824A1 (en) Steering control method and apparatus
CN111959500A (en) Automobile path tracking performance improving method based on tire force distribution
Milani et al. Smart autodriver algorithm for real-time autonomous vehicle trajectory control
Hang et al. Design of an active collision avoidance system for a 4WIS-4WID electric vehicle
Tseng et al. Evasive manoeuvres with a steering robot
CN107856733B (en) A kind of automobile towards man-machine harmony hides dynamic barrier control method
CN107878453B (en) A kind of automobile emergency collision avoidance integral type control method for hiding dynamic barrier
CN113183953A (en) Vehicle post-collision active safety control method and system based on distributed driving chassis
US20230174147A1 (en) Method and control circuit for controlling an active rear axle steering of a motor vehicle when steering out from straight travel, and a motor vehicle having the control circuit
CN107885931B (en) Automobile emergency collision avoidance control method for humanized adjustment of steering wheel sudden change torque
CN116834754A (en) Transverse and longitudinal cooperative control method for self-adaptive speed regulation of automatic driving vehicle

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
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20201009

Termination date: 20211107