CN117111475A - Multi-agent-based active suspension output consistency control method - Google Patents
Multi-agent-based active suspension output consistency control method Download PDFInfo
- Publication number
- CN117111475A CN117111475A CN202311199426.XA CN202311199426A CN117111475A CN 117111475 A CN117111475 A CN 117111475A CN 202311199426 A CN202311199426 A CN 202311199426A CN 117111475 A CN117111475 A CN 117111475A
- Authority
- CN
- China
- Prior art keywords
- output
- vehicle
- controller
- agent
- force
- 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
- 239000000725 suspension Substances 0.000 title claims abstract description 133
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000008878 coupling Effects 0.000 claims abstract description 45
- 238000010168 coupling process Methods 0.000 claims abstract description 45
- 238000005859 coupling reaction Methods 0.000 claims abstract description 45
- 230000001360 synchronised effect Effects 0.000 claims abstract description 26
- 239000003795 chemical substances by application Substances 0.000 claims description 150
- 239000011159 matrix material Substances 0.000 claims description 51
- 238000013016 damping Methods 0.000 claims description 22
- 238000006073 displacement reaction Methods 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 16
- 238000011217 control strategy Methods 0.000 claims description 15
- 238000013461 design Methods 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000005096 rolling process Methods 0.000 claims description 7
- 230000001133 acceleration Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000009467 reduction Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 6
- 239000006096 absorbing agent Substances 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000035939 shock Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006866 deterioration Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60G—VEHICLE SUSPENSION ARRANGEMENTS
- B60G17/00—Resilient suspensions having means for adjusting the spring or vibration-damper characteristics, for regulating the distance between a supporting surface and a sprung part of vehicle or for locking suspension during use to meet varying vehicular or surface conditions, e.g. due to speed or load
- B60G17/015—Resilient suspensions having means for adjusting the spring or vibration-damper characteristics, for regulating the distance between a supporting surface and a sprung part of vehicle or for locking suspension during use to meet varying vehicular or surface conditions, e.g. due to speed or load the regulating means comprising electric or electronic elements
- B60G17/018—Resilient suspensions having means for adjusting the spring or vibration-damper characteristics, for regulating the distance between a supporting surface and a sprung part of vehicle or for locking suspension during use to meet varying vehicular or surface conditions, e.g. due to speed or load the regulating means comprising electric or electronic elements characterised by the use of a specific signal treatment or control method
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Vehicle Body Suspensions (AREA)
Abstract
The invention discloses a multi-agent-based active suspension output consistency control method, which comprises the following steps: establishing a whole vehicle dynamics model and a tire dynamics model; establishing a whole vehicle attitude control model and a vehicle vibration coupling model; designing a vehicle attitude controller and determining a vibration decoupling performance index of the whole vehicle; obtaining expected output of the active suspension system under different driving road conditions; establishing an active suspension system dynamics model based on multiple agents; and the obtained expected output of the active suspension is eliminated through a force synchronous controller, so that the consistency of the output of the active suspension system is ensured. On the basis of ensuring the running stability and safety of the vehicle, the invention weakens the mutual influence of the vibration of the suspension system in all directions, is beneficial to the vibration reduction performance and improves the comfort of the vehicle.
Description
Technical Field
The invention belongs to the technical field of active suspension system control, and particularly relates to an active suspension output consistency control method based on multiple intelligent agents.
Background
As an important component of the vehicle running system, the performance of the suspension system directly affects the stability and safety of the vehicle. The vehicle carrying the active suspension system can control the posture of the whole vehicle in real time according to the topographic information, so that the smoothness, riding comfort and roll stability of the vehicle are improved, and the running safety of the vehicle is further improved.
In the prior art of vehicle active suspension system control, the patent application number of China is CN202011120014.9, which discloses a top-layer vehicle body posture control method based on pitch and roll force compensation, wherein a suspension controller is used for judging whether an ideal damping force is in an effective damping range provided by a shock absorber, the ideal damping force exceeding the effective damping range of the shock absorber is corrected, a logic threshold value control method is used for judging the state of acceleration, and on the basis, the distribution calculation of a pitch force weight coefficient and a roll force weight coefficient is carried out, so that the ideal damping force is always in the effective range provided by the damping force of the shock absorber. The problem of vehicle body posture deterioration in the running process of the vehicle is solved, and riding comfort and safety are improved. The invention discloses an active suspension control method and system, wherein the application number of the active suspension control method is CN201911197599.1, and the method comprises the steps of obtaining a vehicle body attitude expected value by observing a driving road and estimating the self-attitude which is required to be kept by the smoothness of a vehicle, combining with the current vehicle attitude state, calculating to obtain a suspension attitude instruction, and controlling an actuating mechanism of each suspension by a chassis controller to independently adjust, so that the vehicle can adapt to the fluctuation of a road surface, maintain good smoothness of the whole vehicle and improve the riding comfort of the vehicle.
However, the above-mentioned patent technologies do not consider the problem of multi-directional vibration coupling when the vehicle is running and the problem of output inconsistency between suspension subsystems. When the vehicle runs on a road, the generated vibrations in multiple directions have coupling behaviors, and the vibrations in all directions affect each other, so that the driving comfort and the driving safety of the vehicle are not improved. On the other hand, in the control process of the active suspension system of the vehicle, the desired output of the suspension cannot be achieved at the same time due to asynchronous response between the respective suspension actuator actuators. The active suspension system has inconsistent force output, so that the smoothness and riding comfort of the whole vehicle are affected, and the stability and safety of the vehicle are affected to a certain extent. Therefore, in further research and development, the coupling behavior of the vehicle body vibration needs to be analyzed, meanwhile, asynchronous response among suspension executing mechanisms is considered, and corresponding measures are taken to solve the problem of inconsistent force output, so that more excellent vehicle smoothness and riding comfort are realized, and the stability and safety of the vehicle are further improved.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, the present invention is directed to providing a multi-agent-based active suspension output consistency control method, so as to solve the problem of multi-directional vibration coupling and inconsistent output between suspension subsystems in the prior art. The invention provides a layered vehicle body posture control strategy, wherein an upper layer performs decoupling analysis on vertical vibration, pitching vibration and rolling vibration of a vehicle, and a whole vehicle posture controller is designed; the lower layer is based on a multi-agent theory, the vehicle active suspension system is used as a multi-agent system, and by designing a proper multi-agent system consistency control protocol, the active suspension system can ensure that the final state is consistent, and meanwhile, the optimal performance index of the system is realized, so that the stability and the safety of the vehicle during running are improved.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the invention discloses a multi-agent-based active suspension output consistency control method, which comprises the following steps:
(1) Establishing a whole vehicle dynamics model and a tire dynamics model;
(2) Establishing a whole vehicle attitude control model and a vehicle vibration coupling model by utilizing the whole vehicle dynamics model and the tire dynamics model established in the step (1);
(3) Designing a vehicle attitude controller by utilizing the whole vehicle attitude control model and the vehicle vibration coupling model established in the step (2); decoupling analysis is carried out on vertical vibration, pitching vibration and rolling vibration of the vehicle according to the vehicle vibration coupling model, and a decoupling performance index of the vibration of the whole vehicle is determined; the vibration decoupling performance index of the whole vehicle and the expected attitude performance index of the vehicle form a controller optimization target, and the expected output of the active suspension system under different driving road conditions is obtained by solving the optimization target in real time;
(4) Establishing an active suspension system dynamics model based on multiple agents;
(5) Designing an output consistency controller by utilizing the dynamic model of the active suspension system established in the step (4) and the vehicle attitude controller designed in the step (3); the output consistency controller comprises a force tracking controller and a force synchronous controller, the force tracking controller tracks the expected output of the active suspension obtained by solving the vehicle attitude controller designed in the step (3), and the force synchronous controller eliminates inconsistent errors among suspension subsystems, so that the consistency of the output of the active suspension system is ensured.
Further, the specific steps of establishing the whole vehicle dynamics model and the tire dynamics model in the step (1) are as follows:
(11) And (3) a whole vehicle dynamics model:
neglecting the lateral wind and the longitudinal movement of the vehicle during running, only considering the front wheel steering, and obtaining the motion equation of each degree of freedom in the whole vehicle dynamics model by the Darby principle as follows:
the body lateral motion equation is:
wherein m is the mass of the whole vehicle; m is m s Is a sprung mass; u is the speed of the vehicle; beta is the centroid slip angle; omega r Is yaw rate; h is the roll center height; phi is the roll angle of the vehicle body; f (F) y1 ,F y2 ,F y3 ,F y4 The lateral bias forces of the left front tire, the right front tire, the left rear tire and the right rear tire are respectively;
the vehicle yaw motion equation is:
wherein I is z The yaw moment of inertia of the whole vehicle; a, b are distances from the mass center of the vehicle to the front axle and the rear axle respectively;
the roll motion equation of the vehicle body is as follows:
wherein I is x The moment of inertia is the roll of the vehicle body; phi (phi) r Is the road surface transverse gradient angle; f (F) s1 ,F s2 ,F s3 ,F s4 The vertical forces of the left front suspension, the right front suspension, the left rear suspension and the right rear suspension acting on the vehicle body are respectively; t is the track of the vehicle;
the pitching motion equation of the vehicle body is as follows:
wherein I is y The pitching moment of inertia of the vehicle body; θ is the pitch angle of the vehicle body;
The vertical equation of motion at the body centroid is:
wherein x is s The center of mass of the sprung mass is displaced vertically;
(12) Tire dynamics model:
ignoring the effect of tire characteristic change and tire aligning moment caused by load change, the tire characteristic is linear under the condition of small rotation angle, and the cornering force of the tire is expressed as:
F 1 ＝F 2 ＝k 1 α 1 (7)
F 3 ＝F 4 ＝k 2 α 2 (8)
wherein delta is the steering angle of the front wheel; k (k) 1 And k 2 The lateral deflection rigidity of the front wheel and the rear wheel is respectively; alpha 1 And alpha 2 The front wheel and the rear wheel are respectively provided with a side deflection angle; e (E) f And E is r The front wheel and the rear wheel are respectively provided with side steering coefficients;
the vertical load of the tire is expressed as:
wherein F is z1 ，F z2 ，F z3 ，F z4 Vertical loads of the left front tire, the right front tire, the left rear tire, and the right rear tire, respectively; k (k) t1 ，k t2 ，k t3 ，k t4 The rigidity of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x t1 ，x t2 ，x t3 ，x t4 Inputting excitation to the road surfaces of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x us1 ，x us2 ，x us3 ，x us4 Respectively a left front tire,The unsprung mass at the front right tire, rear left tire, and rear right tire is displaced vertically.
Further, the specific steps of establishing the whole vehicle attitude control model and the vehicle vibration coupling model in the step (2) are as follows:
(21) Establishing a whole vehicle attitude control model;
(211) The method comprises the following steps of establishing a prediction model of the whole vehicle system;
And (3) utilizing the dynamics model established in the steps (11) and (12) to establish a whole vehicle system prediction model as follows:
the prediction model of the whole vehicle system is expressed by a state space, and the state variable x of the system is taken as follows:
the control input u is:
u＝[F s1 ,F s2 ,F s3 ,F s4 ] T
the control output y is:
y＝[ω r ,β,φ,θ,x s ] T
the state space of the prediction model of the whole vehicle system is expressed as follows:
wherein A is u An 8 x 8 dimensional system matrix; b (B) u Control input matrix for 8 x 4 dimensions; c (C) u Output a state matrix for 5 x 8 dimensions; each matrix can be obtained by solving according to a whole vehicle system prediction model;
(212) Establishing a whole vehicle attitude control model;
the method of a first-order difference quotient is applied to discretize the state space of a prediction model of the whole vehicle system, and a discretized state space equation is expressed as:
wherein,
wherein I represents a group A u Identity matrix of the same matrix dimension, A k ，B k ，C k Respectively corresponding to the discretized coefficient matrixes; t is t s Is discretized time;
(22) The vehicle vibration coupling model is established specifically as follows:
the vehicle vibration coupling model is built as follows:
the vertical displacement of the sprung mass at the suspension is expressed as follows: :
wherein x is s1 ，x s2 ，x s3 ，x s4 Vertical displacements of the left front sprung mass, the right front sprung mass, the left rear sprung mass and the right rear sprung mass, respectively;
the output force of the suspension system is expressed as follows:
Combining equation (14) and equation (15), the vehicle vibration coupling model usable state space is expressed as follows:
in the method, in the process of the invention,
wherein k is 1 ,k 2 ,k 3 ,k 4 Equivalent elastic coefficients, c, of the 4 suspension subsystems, respectively 1 ,c 2 ,c 3 ,c 4 Equivalent damping coefficients of the 4 suspension subsystems respectively; a is that s Damping matrix for vibration coupling model of vehicle, B s A rigidity matrix of the vibration coupling model of the vehicle; the above mentioned classesThe effective elastic coefficient and the damping coefficient can be obtained by solving in real time through a system identification algorithm, and the specific method comprises, but is not limited to, a least square method and Kalman filtering.
Further, the design steps of the vehicle posture controller in the step (3) are as follows:
(31) Vehicle vibration decoupling analysis;
decoupling analysis is carried out on the vertical vibration, the pitching vibration and the rolling vibration of the vehicle according to the vehicle vibration coupling model established in the step (22), and a damping matrix A of the formula (16) is obtained s And a stiffness matrix B s The off-diagonal element of (2) is not zero, there is a coupling behavior of the vehicle vertical vibration, pitch vibration and roll vibration, and when the off-diagonal element is zero, the coupling term is not present, and vibrations in various directions of the vehicle are no longer coupled, as shown below:
[a 2 a 3 a 4 a 6 a 7 a 8 b 2 b 3 b 4 b 6 b 7 b 8 ] T ＝0 12×1 (17)
bringing the specific expression in the formula (16) into the formula (17) can be simplified as follows:
(32) Determining the vibration decoupling performance index of the whole vehicle;
The vibration coupling behavior of the vehicle is reduced by optimizing the vibration decoupling performance index of the whole vehicle, and the vibration decoupling performance index eta of the whole vehicle is expressed as follows:
(33) Determining the input and output of a controller;
taking the real-time detected vehicle body state quantity as the input of a controller, and taking the expected output force of each subsystem in the active suspension system as the output of the controller;
(34) Determining a cost function;
determining a cost function of the controller according to the vehicle body posture performance index, wherein factors influencing the vehicle body posture performance index comprise: vehicle vertical acceleration, roll angle, pitch angle and tire dynamic displacement; and combining the vibration decoupling performance index of the whole vehicle determined in the step (32), and designing a cost function of the controller as follows:
wherein J is k Represents the controller cost function value, Q 1 ，Q 2 ，Q 3 ，Q 4 ，Q 5 A weight matrix representing a state quantity, R represents a weight matrix of a control quantity, P h Representing the prediction time domain, C h Represents the control time domain, the relaxation factor ε= [ ε ] 1 ,ε 2 ,ε 3 ,ε 4 ] T ρ is the weight coefficient of the relaxation factor;
(35) Solving a system cost function;
and (3) obtaining a required control output by solving the following optimal problem by using the cost function in the whole vehicle attitude control model step (34) in the step (21):
s.t.u min ≤u(k)≤u max
The optimization problem is converted into a quadratic programming problem, and the quadratic programming problem is solved through an effective set method or an interior point method, so that a series of output variables of the controller in a control time domain are obtained:
u * ＝[u * (k),u * (k+1),u * (k+2),…,u * (k+C h -1)]
taking the first element of the control sequence as an actual output variable of the controller, and performing control on the active suspension system until the next discretization time; the controller predicts the state of the next discretization time according to the whole vehicle attitude control model, and solves and generates a new control output variable sequence again; and the real-time expected output of the active suspension system under different driving road conditions is obtained by circulating in this way.
Further, the specific steps of establishing the dynamic model of the active suspension system based on the multiple agents in the step (4) are as follows:
(41) Active suspension system dynamics model based on multiple agents: taking 4 suspension subsystems of an active suspension system as 4 agents, wherein the agents are mutually communicated, and the dynamics equation of the ith agent is as follows:
wherein F is si Output force for the i-th agent, i=1, 2,3,4; k (k) si Suspension stiffness coefficient for the ith agent; c si Suspension damping coefficient for the ith agent; f (f) i Actuator force for the ith agent; x is x si Is the displacement at the ith agent tire;
Vertical motion of the ith agent:
wherein m is usi Is the mass of the ith agent; x is x usi Is the vertical displacement of the ith agent; f (F) zi Representing the total vertical load of the ith agent; m is m si Representing the suspended mass of the ith agent; θ r Representing the road surface longitudinal gradient angle; f (F) zi Is the vertical load on the ith agent tire;
(42) State space equation of the agent: based on the dynamics model established in the step (41), the state space equation of the ith agent is obtained as follows:
in the method, in the process of the invention,u i ＝[f i ]，y i ＝[F si ]the state variable, the control input variable and the system output variable of the ith agent are respectively; a is that i ，B i ，C i ，D i ，E i Respectively representing corresponding coefficient matrixes, and solving according to the dynamics model established in the step (41); d, d i Indicating the road surface stimulus to which the i-th agent is subjected.
Further, the specific steps of designing the output consistency controller in the step (5) are as follows:
the communication topological property among the intelligent agents is represented by graph theory knowledge, and the active suspension system is an undirected graph;
undirected graph g= { R, E } is composed of 4 vertex sets R 4 = {1,2,3,4} and edge set E; the weighted adjacency matrix of undirected graph G is denoted as a= [ a ] mn ]When edge (m, n) ∈E, a mn If the value is more than 0, the node m and the node n are called neighbor nodes, and the state information of each other is obtained; the edge set e= { (1, 2), (1, 3), (1, 4), (2, 1), (2, 3), (2, 4), (3, 1), (3, 2), (3, 4), (4, 1), (4, 2), (4, 3) };
(51) The design consistency control strategy is as follows:
in consideration of the requirement of output consistency among suspension subsystems, an active suspension system output consistency control strategy is designed based on a multi-agent theory, and the final output of an agent i consists of two parts, namely:
f i_ref ＝f i_refd +f i_refm (24)
wherein f i_ref The total output of the output consistency controller of the intelligent agent i; f (f) i_refd The method comprises the steps that the output of a force tracking controller of an intelligent agent i is obtained through solving of the force tracking controller; f (f) i_refm The method comprises the steps that the output of a force synchronous controller of an intelligent agent i is obtained through solving of the force synchronous controller;
calculating the average value of the output forces of all the agents, taking the difference between the output force of each agent and the average value to obtain a deviation value, and taking the deviation value of each agent as the input of the force synchronous controller; normalizing the output force of the intelligent agent, and calculating by adopting the normalized output force after the normalization;
the average value calculation formula is as follows:
wherein F is m The average value of the output force of the active suspension system is obtained; f (F) i Representing the actual output force of the ith agent; f (F) i_ref Representing the expected output force of the ith agent;
the calculation formula of the deviation value of each agent is as follows:
in the formula e i_m Representing the deviation value of the ith agent;
defining the output consistency error of the intelligent agent i to embody the output consistency error among the intelligent agents, wherein the method comprises the following steps of:
Wherein a is ij ∈A，H i Is a weight coefficient matrix;
the multi-agent system consistency control strategy is designed as follows:
in the formula e i_d ＝F i_ref -F i Representing a force tracking error of the ith agent; I.I.I. represents a matrix or the 2 norms of the vectors;
the output consistency of the multi-agent system can be ensured by solving the expected output of the agents meeting the strategy;
(52) Designing a force tracking controller:
solving the vehicle attitude controller designed in the step (3) to obtain the expected output force of the active suspension system, wherein the expected output force is F respectively s1 ,F s2 ,F s3 ,F s4 Taking the expected output force of the suspension subsystem as the expected output force of the corresponding agent:
F i_ref ＝F si (29)
the design control cost function is as follows:
wherein J is i_d Representing a force tracking controller control cost function value; p (P) hid Representing a force tracking controller prediction horizon; c (C) hid Representing a force tracking controller control time domain; q (Q) id1 ,Q id2 ,Q id3 Respectively representing state quantity weight matrixes of the intelligent agents i; r is R id Representing the control of the output weight matrix by the force tracking controller;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical tracking controller i_refd ；
(53) Designing a force synchronization controller:
according to the consistency control strategy designed in the step (51), the design force synchronous controller controls the cost function as follows:
wherein J is i_m Representing the value of the cost function controlled by the force synchronization controller; p (P) him Representing a predicted time domain of the force synchronization controller; c (C) him Representing the force synchronization controller control time domain; q (Q) im1 ,Q im2 ,Q im3 ,Q im4 Respectively representing related state quantity weight matrixes; r is R im The representation force synchronous controller controls the output weight matrix;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical force synchronous controller i_refm 。
The invention has the beneficial effects that:
1. the invention provides a layered vehicle body posture control strategy, which improves the comfort, the running stability and the safety of the vehicle;
2. aiming at the problem of multidirectional vibration coupling in the prior art, an upper-layer vehicle posture controller is designed, a vehicle vibration coupling model is established, decoupling analysis is carried out on vertical vibration, pitching vibration and rolling vibration of a vehicle, a whole vehicle vibration decoupling performance index is determined, and the mutual influence of vibration in all directions of a suspension system is weakened on the basis of ensuring the running stability and safety of the vehicle by combining the whole vehicle vibration decoupling performance index and a vehicle expected posture performance index, so that the vibration damping performance is facilitated, and the comfort of the vehicle is improved;
3. The invention designs a lower layer output consistency controller aiming at the problem of inconsistent output among suspension subsystems in the prior art, treats an active suspension system as a multi-agent system based on a multi-agent theory, designs an active suspension system output consistency control strategy, solves the obtained expected output of the active suspension system according to a vehicle attitude controller, solves the obtained output of a force tracking controller and the output of a force synchronous controller to jointly form the total output of a suspension frame system, ensures the consistency of the output of the active suspension system, realizes the global optimum of system performance indexes, and greatly improves the stability and safety of a vehicle during running.
Drawings
FIG. 1 is a schematic diagram of the method of the present invention.
FIG. 2 is a block diagram of a coherence controller in accordance with the present invention.
Detailed Description
The invention will be further described with reference to examples and drawings, to which reference is made, but which are not intended to limit the scope of the invention.
Referring to fig. 1 and 2, the method for controlling the consistency of the output of the active suspension based on the multiple intelligent agents comprises the following steps:
(1) Establishing a whole vehicle dynamics model and a tire dynamics model; the method comprises the following specific steps:
(11) And (3) a whole vehicle dynamics model:
neglecting the lateral wind and the longitudinal movement of the vehicle during running, only considering the front wheel steering, and obtaining the motion equation of each degree of freedom in the whole vehicle dynamics model by the Darby principle as follows:
the body lateral motion equation is:
wherein m is the mass of the whole vehicle; m is m s Is a sprung mass; u is the speed of the vehicle; beta is the centroid slip angle; omega r Is yaw rate; h is the roll center height; phi is the roll angle of the vehicle body; f (F) y1 ,F y2 ,F y3 ,F y4 The lateral bias forces of the left front tire, the right front tire, the left rear tire and the right rear tire are respectively;
the vehicle yaw motion equation is:
wherein I is z The yaw moment of inertia of the whole vehicle; a, b are distances from the mass center of the vehicle to the front axle and the rear axle respectively;
the roll motion equation of the vehicle body is as follows:
wherein I is x The moment of inertia is the roll of the vehicle body; phi (phi) r Is the road surface transverse gradient angle; f (F) s1 ,F s2 ,F s3 ,F s4 The vertical forces of the left front suspension, the right front suspension, the left rear suspension and the right rear suspension acting on the vehicle body are respectively; t is the track of the vehicle;
the pitching motion equation of the vehicle body is as follows:
wherein I is y The pitching moment of inertia of the vehicle body; θ is the pitch angle of the vehicle body;
the vertical equation of motion at the body centroid is:
Wherein x is s The center of mass of the sprung mass is displaced vertically;
(12) Tire dynamics model:
ignoring the effect of tire characteristic change and tire aligning moment caused by load change, the tire characteristic is linear under the condition of small rotation angle, and the cornering force of the tire is expressed as:
F 1 ＝F 2 ＝k 1 α 1 (7)
F 3 ＝F 4 ＝k 2 α 2 (8)
wherein delta is the steering angle of the front wheel; k (k) 1 And k 2 The lateral deflection rigidity of the front wheel and the rear wheel is respectively; alpha 1 And alpha 2 The front wheel and the rear wheel are respectively provided with a side deflection angle; e (E) f And E is r The front wheel and the rear wheel are respectively provided with side steering coefficients;
the vertical load of the tire is expressed as:
wherein F is z1 ，F z2 ，F z3 ，F z4 Vertical loads of the left front tire, the right front tire, the left rear tire, and the right rear tire, respectively; k (k) t1 ，k t2 ，k t3 ，k t4 The rigidity of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x t1 ，x t2 ，x t3 ，x t4 Inputting excitation to the road surfaces of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x us1 ，x us2 ，x us3 ，x us4 The unsprung mass at the left front tire, right front tire, left rear tire, and right rear tire, respectively, are displaced vertically.
(2) Establishing a whole vehicle attitude control model and a vehicle vibration coupling model by utilizing the whole vehicle dynamics model and the tire dynamics model established in the step (1); the method comprises the following specific steps:
(21) Establishing a whole vehicle attitude control model;
(211) The method comprises the following steps of establishing a prediction model of the whole vehicle system;
and (3) utilizing the dynamics model established in the steps (11) and (12) to establish a whole vehicle system prediction model as follows:
the prediction model of the whole vehicle system is expressed by a state space, and the state variable x of the system is taken as follows:
the control input u is:
u＝[F s1 ,F s2 ,F s3 ,F s4 ] T
the control output y is:
y＝[ω r ,β,φ,θ,x s ] T
the state space of the prediction model of the whole vehicle system is expressed as follows:
wherein A is u An 8 x 8 dimensional system matrix; b (B) u Control input matrix for 8 x 4 dimensions; c (C) u Output a state matrix for 5 x 8 dimensions; each matrix can be obtained by solving according to a whole vehicle system prediction model;
(212) Establishing a whole vehicle attitude control model;
the method of a first-order difference quotient is applied to discretize the state space of a prediction model of the whole vehicle system, and a discretized state space equation is expressed as:
wherein,
wherein I represents a group A u Identity matrix of the same matrix dimension, A k ，B k ，C k Respectively corresponding to the discretized coefficient matrixes;t s is discretized time;
(22) The vehicle vibration coupling model is established specifically as follows:
the vehicle vibration coupling model is built as follows:
the vertical displacement of the sprung mass at the suspension is expressed as follows: :
wherein x is s1 ，x s2 ，x s3 ，x s4 Vertical displacements of the left front sprung mass, the right front sprung mass, the left rear sprung mass and the right rear sprung mass, respectively;
The output force of the suspension system is expressed as follows:
combining equation (14) and equation (15), the vehicle vibration coupling model usable state space is expressed as follows:
in the method, in the process of the invention,
wherein k is 1 ,k 2 ,k 3 ,k 4 Equivalent elastic coefficients, c, of the 4 suspension subsystems, respectively 1 ,c 2 ,c 3 ,c 4 Equivalent damping coefficients of the 4 suspension subsystems respectively; a is that s Damping matrix for vibration coupling model of vehicle, B s A rigidity matrix of the vibration coupling model of the vehicle; the equivalent elastic coefficient and the damping coefficient can be obtained by solving in real time through a system identification algorithm, and the specific method comprises, but is not limited to, a least square method and a Kalman filter.
(3) Designing a vehicle attitude controller by utilizing the whole vehicle attitude control model and the vehicle vibration coupling model established in the step (2); decoupling analysis is carried out on vertical vibration, pitching vibration and rolling vibration of the vehicle according to the vehicle vibration coupling model, and a decoupling performance index of the vibration of the whole vehicle is determined; the vibration decoupling performance index of the whole vehicle and the expected attitude performance index of the vehicle form a controller optimization target, and the expected output of the active suspension system under different driving road conditions is obtained by solving the optimization target in real time;
the design steps of the vehicle attitude controller are as follows:
(31) Vehicle vibration decoupling analysis;
Decoupling analysis is carried out on the vertical vibration, the pitching vibration and the rolling vibration of the vehicle according to the vehicle vibration coupling model established in the step (22), and a damping matrix A of the formula (16) is obtained s And a stiffness matrix B s The off-diagonal element of (2) is not zero, there is a coupling behavior of the vehicle vertical vibration, pitch vibration and roll vibration, and when the off-diagonal element is zero, the coupling term is not present, and vibrations in various directions of the vehicle are no longer coupled, as shown below:
[a 2 a 3 a 4 a 6 a 7 a 8 b 2 b 3 b 4 b 6 b 7 b 8 ] T ＝0 12×1 (17)
bringing the specific expression in the formula (16) into the formula (17) can be simplified as follows:
(32) Determining the vibration decoupling performance index of the whole vehicle;
the vibration coupling behavior of the vehicle is reduced by optimizing the vibration decoupling performance index of the whole vehicle, and the vibration decoupling performance index eta of the whole vehicle is expressed as follows:
(33) Determining the input and output of a controller;
taking the real-time detected vehicle body state quantity as the input of a controller, and taking the expected output force of each subsystem in the active suspension system as the output of the controller;
(34) Determining a cost function;
determining a cost function of the controller according to the vehicle body posture performance index, wherein factors influencing the vehicle body posture performance index comprise: vehicle vertical acceleration, roll angle, pitch angle and tire dynamic displacement; and combining the vibration decoupling performance index of the whole vehicle determined in the step (32), and designing a cost function of the controller as follows:
Wherein J is k Represents the controller cost function value, Q 1 ，Q 2 ，Q 3 ，Q 4 ，Q 5 A weight matrix representing a state quantity, R represents a weight matrix of a control quantity, P h Representing the prediction time domain, C h Representing the control time domain, since the system model is changed in real time and the optimization objective cannot be guaranteed to obtain a feasible solution at each moment, a relaxation factor is added into the cost function, and the relaxation factor epsilon= [ epsilon ] 1 ,ε 2 ,ε 3 ,ε 4 ] T ρ is the weight coefficient of the relaxation factor;
(35) Solving a system cost function;
and (3) obtaining a required control output by solving the following optimal problem by using the cost function in the whole vehicle attitude control model step (34) in the step (21):
s.t.u min ≤u(k)≤u max
the optimization problem is converted into a quadratic programming problem, and the quadratic programming problem is solved through an effective set method or an interior point method, so that a series of output variables of the controller in a control time domain are obtained:
u * ＝[u * (k),u * (k+1),u * (k+2),…,u * (k+C h -1)]
taking the first element of the control sequence as an actual output variable of the controller, and performing control on the active suspension system until the next discretization time; the controller predicts the state of the next discretization time according to the whole vehicle attitude control model, and solves and generates a new control output variable sequence again; and the real-time expected output of the active suspension system under different driving road conditions is obtained by circulating in this way.
(4) Establishing an active suspension system dynamics model based on multiple agents; the method comprises the following specific steps:
(41) Active suspension system dynamics model based on multiple agents: taking 4 suspension subsystems of an active suspension system as 4 agents, wherein the agents are mutually communicated, and the dynamics equation of the ith agent is as follows:
wherein F is si Output force for the i-th agent, i=1, 2,3,4; k (k) si Suspension stiffness coefficient for the ith agent; c si Suspension damping coefficient for the ith agent; f (f) i Actuator force for the ith agent; x is x si Is the displacement at the ith agent tire;
vertical motion of the ith agent:
wherein m is usi Is the mass of the ith agent; x is x usi Is the vertical displacement of the ith agent; f (F) zi Representing the total vertical load of the ith agent; m is m si Representing the suspended mass of the ith agent; θ r Representing the road surface longitudinal gradient angle; f (F) zi Is the vertical load on the ith agent tire;
(42) State space equation of the agent: based on the dynamics model established in the step (41), the state space equation of the ith agent is obtained as follows:
in the method, in the process of the invention,u i ＝[f i ]，y i ＝[F si ]the state variable, the control input variable and the system output variable of the ith agent are respectively; a is that i ，B i ，C i ，D i ，E i Respectively representing corresponding coefficient matrixes, and solving according to the dynamics model established in the step (41); d, d i Indicating the road surface stimulus to which the i-th agent is subjected.
(5) Designing an output consistency controller by utilizing the dynamic model of the active suspension system established in the step (4) and the vehicle attitude controller designed in the step (3); the output consistency controller comprises a force tracking controller and a force synchronous controller, the force tracking controller tracks the expected output of the active suspension obtained by solving the vehicle attitude controller designed in the step (3), and the force synchronous controller eliminates inconsistent errors among suspension subsystems, so that the consistency of the output of the active suspension system is ensured;
the specific steps of designing the output consistency controller in the step (5) are as follows:
the communication topological property among the intelligent agents is represented by graph theory knowledge, and the active suspension system is an undirected graph;
undirected graph g= { R, E } is composed of 4 vertex sets R 4 = {1,2,3,4} and edge set E; the weighted adjacency matrix of undirected graph G is denoted as a= [ a ] mn ]When edge (m, n) ∈E, a mn If the value is more than 0, the node m and the node n are called neighbor nodes, and the state information of each other is obtained; the edge set e= { (1, 2), (1, 3), (1, 4), (2, 1), (2, 3), (2, 4), (3, 1), (3, 2), (3, 4), (4, 1), (4, 2), (4, 3) };
(51) The design consistency control strategy is as follows:
in consideration of the requirement of output consistency among suspension subsystems, an active suspension system output consistency control strategy is designed based on a multi-agent theory, and the final output of an agent i consists of two parts, namely:
f i_ref ＝f i_refd +f i_refm (24)
wherein f i_ref The total output of the output consistency controller of the intelligent agent i; f (f) i_refd The method comprises the steps that the output of a force tracking controller of an intelligent agent i is obtained through solving of the force tracking controller; f (f) i_refm The method comprises the steps that the output of a force synchronous controller of an intelligent agent i is obtained through solving of the force synchronous controller;
calculating the average value of the output forces of all the agents, taking the difference between the output force of each agent and the average value to obtain a deviation value, and taking the deviation value of each agent as the input of the force synchronous controller; because the deviation value is obtained by the average value of the output force of all the agents and the difference value of the output force of each agent, the deviation value connects all the agents together, so that the influence of each agent in the system on other agents is minimized, and the output consistency of the system is ensured to be optimal; because the expected output force of each intelligent body is different, normalizing the output force of the intelligent body, and calculating by adopting the normalized output force after the normalization;
The average value calculation formula is as follows:
wherein F is m The average value of the output force of the active suspension system is obtained; f (F) i Representing the actual output force of the ith agent; f (F) i_ref Representing the expected output force of the ith agent;
the calculation formula of the deviation value of each agent is as follows:
in the formula e i_m Representing the deviation value of the ith agent;
defining the output consistency error of the intelligent agent i to embody the output consistency error among the intelligent agents, wherein the method comprises the following steps of:
wherein a is ij ∈A，H i Is a weight coefficient matrix;
in the defined output consistency errors, besides the tracking error of the intelligent agent i, the output error between the intelligent agent i and the intelligent agent j also occurs; therefore, even if the expected output of each suspension intelligent body is different, the expected output can be achieved at the same time, and the expected posture of the vehicle is ensured;
the multi-agent system consistency control strategy is designed as follows:
in the formula e i_d ＝F i_ref -F i Representing a force tracking error of the ith agent; II indicates the 2 norms of the matrix or vector;
the output consistency of the multi-agent system can be ensured by solving the expected output of the agents meeting the strategy;
(52) Designing a force tracking controller:
solving the vehicle attitude controller designed in the step (3) to obtain the expected output force of the active suspension system, wherein the expected output force is F respectively s1 ,F s2 ,F s3 ,F s4 Taking the expected output force of the suspension subsystem as the expected output force of the corresponding agent:
F i_ref ＝F si (29)
the main objective of the force tracking controller is to control the actual output force of the intelligent body to track the expected output force in real time, so as to ensure the tracking precision of the system, and simultaneously, in order to ensure the driving comfort, the vertical speed and acceleration of the suspension system need to be controlled as small as possible, therefore, the design control cost function is as follows:
wherein J is i_d Representing a force tracking controller control cost function value; p (P) hid Representing a force tracking controller prediction horizon; c (C) hid Representing a force tracking controller control time domain; q (Q) id1 ,Q id2 ,Q id3 Respectively representing state quantity weight matrixes of the intelligent agents i; r is R id Representing the control of the output weight matrix by the force tracking controller;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical tracking controller i_refd ；
(53) Designing a force synchronization controller:
according to the consistency control strategy designed in the step (51), the design force synchronous controller controls the cost function as follows:
Wherein J is i_m Representing the value of the cost function controlled by the force synchronization controller; p (P) him Representing a predicted time domain of the force synchronization controller; c (C) him Representing the force synchronization controller control time domain; q (Q) im1 ,Q im2 ,Q im3 ,Q im4 Respectively representing related state quantity weight matrixes; r is R im The representation force synchronous controller controls the output weight matrix;
the cost function according to the design can realize consistency among force output, displacement output, speed output and acceleration output of the suspension subsystems so as to solve the problem of vehicle stability and safety reduction caused by inconsistent output of the active suspension system due to asynchronous response among the execution mechanisms in the prior art;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical force synchronous controller i_refm 。
The present invention has been described in terms of the preferred embodiments thereof, and it should be understood by those skilled in the art that various modifications can be made without departing from the principles of the invention, and such modifications should also be considered as being within the scope of the invention.
Claims (6)
1. The active suspension output consistency control method based on the multiple agents is characterized by comprising the following steps:
(1) Establishing a whole vehicle dynamics model and a tire dynamics model;
(2) Establishing a whole vehicle attitude control model and a vehicle vibration coupling model;
(3) Designing a vehicle attitude controller, and carrying out decoupling analysis on vertical vibration, pitching vibration and rolling vibration of the vehicle according to the vehicle vibration coupling model to determine a decoupling performance index of the vibration of the whole vehicle; the vibration decoupling performance index of the whole vehicle and the expected attitude performance index of the vehicle form a controller optimization target, and the expected output of the active suspension system under different driving road conditions is obtained by solving the optimization target in real time;
(4) Establishing an active suspension system dynamics model based on multiple agents;
(5) Designing an output consistency controller by utilizing the dynamic model of the active suspension system established in the step (4) and the vehicle attitude controller designed in the step (3); the output consistency controller comprises a force tracking controller and a force synchronous controller, the force tracking controller tracks the expected output of the active suspension obtained by solving the vehicle attitude controller designed in the step (3), and the force synchronous controller eliminates inconsistent errors among suspension subsystems, so that the consistency of the output of the active suspension system is ensured.
2. The multi-agent-based active suspension output consistency control method according to claim 1, wherein the specific steps of establishing the whole vehicle dynamics model and the tire dynamics model in the step (1) are as follows:
(11) And (3) a whole vehicle dynamics model:
neglecting the lateral wind and the longitudinal movement of the vehicle during running, only considering the front wheel steering, and obtaining the motion equation of each degree of freedom in the whole vehicle dynamics model by the Darby principle as follows:
the body lateral motion equation is:
wherein m is the mass of the whole vehicle; m is m s Is a sprung mass; u is the speed of the vehicle; beta is the centroid slip angle; omega r Is yaw rate; h is the roll center height; phi is the roll angle of the vehicle body; f (F) y1 ,F y2 ,F y3 ,F y4 The lateral bias forces of the left front tire, the right front tire, the left rear tire and the right rear tire are respectively;
the vehicle yaw motion equation is:
wherein I is z The yaw moment of inertia of the whole vehicle; a, b are distances from the mass center of the vehicle to the front axle and the rear axle respectively;
the roll motion equation of the vehicle body is as follows:
wherein I is x The moment of inertia is the roll of the vehicle body; phi (phi) r Is the road surface transverse gradient angle; f (F) s1 ,F s2 ,F s3 ,F s4 The vertical forces of the left front suspension, the right front suspension, the left rear suspension and the right rear suspension acting on the vehicle body are respectively; t is the track of the vehicle;
The pitching motion equation of the vehicle body is as follows:
wherein I is y The pitching moment of inertia of the vehicle body; θ is the pitch angle of the vehicle body;
the vertical equation of motion at the body centroid is:
wherein x is s The center of mass of the sprung mass is displaced vertically;
(12) Tire dynamics model:
ignoring the effect of tire characteristic change and tire aligning moment caused by load change, the tire characteristic is linear under the condition of small rotation angle, and the cornering force of the tire is expressed as:
F 1 ＝F 2 ＝k 1 α 1 (7)
F 3 ＝F 4 ＝k 2 α 2 (8)
wherein delta is the steering angle of the front wheel; k (k) 1 And k 2 The lateral deflection rigidity of the front wheel and the rear wheel is respectively;α 1 and alpha 2 The front wheel and the rear wheel are respectively provided with a side deflection angle; e (E) f And E is r The front wheel and the rear wheel are respectively provided with side steering coefficients;
the vertical load of the tire is expressed as:
wherein F is z1 ，F z2 ，F z3 ，F z4 Vertical loads of the left front tire, the right front tire, the left rear tire, and the right rear tire, respectively; k (k) t1 ，k t2 ，k t3 ，k t4 The rigidity of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x t1 ，x t2 ，x t3 ，x t4 Inputting excitation to the road surfaces of the left front tire, the right front tire, the left rear tire and the right rear tire respectively; x is x us1 ，x us2 ，x us3 ，x us4 The unsprung mass at the left front tire, right front tire, left rear tire, and right rear tire, respectively, are displaced vertically.
3. The multi-agent-based active suspension output consistency control method according to claim 2, wherein the specific steps of establishing the whole vehicle attitude control model and the vehicle vibration coupling model in the step (2) are as follows:
(21) Establishing a whole vehicle attitude control model;
(211) The method comprises the following steps of establishing a prediction model of the whole vehicle system;
and (3) utilizing the dynamics model established in the steps (11) and (12) to establish a whole vehicle system prediction model as follows:
the prediction model of the whole vehicle system is expressed by a state space, and the state variable x of the system is taken as follows:
the control input u is:
u＝[F s1 ,F s2 ,F s3 ,F s4 ] T
the control output y is:
y＝[ω r ,β,φ,θ,x s ] T
the state space of the prediction model of the whole vehicle system is expressed as follows:
wherein A is u An 8 x 8 dimensional system matrix; b (B) u Control input matrix for 8 x 4 dimensions; c (C) u Output a state matrix for 5 x 8 dimensions; each matrix can be obtained by solving according to a whole vehicle system prediction model;
(212) Establishing a whole vehicle attitude control model;
the method of a first-order difference quotient is applied to discretize the state space of a prediction model of the whole vehicle system, and a discretized state space equation is expressed as:
wherein,
wherein I represents a group A u Identity matrix of the same matrix dimension, A k ，B k ，C k Respectively corresponding to the discretized coefficient matrixes; t is t s Is discretized time;
(22) The vehicle vibration coupling model is established specifically as follows:
the vehicle vibration coupling model is built as follows:
the vertical displacement of the sprung mass at the suspension is expressed as follows: :
wherein x is s1 ，x s2 ，x s3 ，x s4 Vertical displacements of the left front sprung mass, the right front sprung mass, the left rear sprung mass and the right rear sprung mass, respectively;
The output force of the suspension system is expressed as follows:
combining equation (14) and equation (15), the vehicle vibration coupling model usable state space is expressed as follows:
in the method, in the process of the invention,
wherein k is 1 ,k 2 ,k 3 ,k 4 Equivalent elastic coefficients, c, of the 4 suspension subsystems, respectively 1 ,c 2 ,c 3 ,c 4 Equivalent damping coefficients of the 4 suspension subsystems respectively; a is that s Damping matrix for vibration coupling model of vehicle, B s A rigidity matrix of the vibration coupling model of the vehicle; the equivalent elastic coefficient and the damping coefficient can be obtained by solving in real time through a system identification algorithm, and the specific method comprises, but is not limited to, a least square method and a Kalman filter.
4. The multi-agent based active suspension output consistency control method according to claim 3, wherein the vehicle attitude controller in step (3) is designed as follows:
(31) Vehicle vibration decoupling analysis;
solving for vertical vibration, pitch vibration and roll vibration of the vehicle according to the vehicle vibration coupling model established in step (22)Coupling analysis is carried out to obtain a damping matrix A in the formula (16) s And a stiffness matrix B s The off-diagonal element of (2) is not zero, there is a coupling behavior of the vehicle vertical vibration, pitch vibration and roll vibration, and when the off-diagonal element is zero, the coupling term is not present, and vibrations in various directions of the vehicle are no longer coupled, as shown below:
[a 2 a 3 a 4 a 6 a 7 a 8 b 2 b 3 b 4 b 6 b 7 b 8 ] T ＝0 12×1 (17)
Bringing the specific expression in the formula (16) into the formula (17) can be simplified as follows:
(32) Determining the vibration decoupling performance index of the whole vehicle;
the vibration coupling behavior of the vehicle is reduced by optimizing the vibration decoupling performance index of the whole vehicle, and the vibration decoupling performance index eta of the whole vehicle is expressed as follows:
(33) Determining the input and output of a controller;
taking the real-time detected vehicle body state quantity as the input of a controller, and taking the expected output force of each subsystem in the active suspension system as the output of the controller;
(34) Determining a cost function;
determining a cost function of the controller according to the vehicle body posture performance index, wherein factors influencing the vehicle body posture performance index comprise: vehicle vertical acceleration, roll angle, pitch angle and tire dynamic displacement; and combining the vibration decoupling performance index of the whole vehicle determined in the step (32), and designing a cost function of the controller as follows:
wherein J is k Represents the controller cost function value, Q 1 ，Q 2 ，Q 3 ，Q 4 ，Q 5 A weight matrix representing a state quantity, R represents a weight matrix of a control quantity, P h Representing the prediction time domain, C h Represents the control time domain, the relaxation factor ε= [ ε ] 1 ,ε 2 ,ε 3 ,ε 4 ] T ρ is the weight coefficient of the relaxation factor;
(35) Solving a system cost function;
and (3) obtaining a required control output by solving the following optimal problem by using the cost function in the whole vehicle attitude control model step (34) in the step (21):
s.t.u min ≤u(k)≤u max
The optimization problem is converted into a quadratic programming problem, and the quadratic programming problem is solved through an effective set method or an interior point method, so that a series of output variables of the controller in a control time domain are obtained:
u * ＝[u * (k),u * (k+1),u * (k+2),…,u * (k+C h -1)]
taking the first element of the control sequence as an actual output variable of the controller, and performing control on the active suspension system until the next discretization time; the controller predicts the state of the next discretization time according to the whole vehicle attitude control model, and solves and generates a new control output variable sequence again; and the real-time expected output of the active suspension system under different driving road conditions is obtained by circulating in this way.
5. The method for controlling the consistency of the output of the active suspension based on the multiple agents according to claim 4, wherein the specific steps of establishing the dynamic model of the active suspension system based on the multiple agents in the step (4) are as follows:
(41) Active suspension system dynamics model based on multiple agents: taking 4 suspension subsystems of an active suspension system as 4 agents, wherein the agents are mutually communicated, and the dynamics equation of the ith agent is as follows:
wherein F is si Output force for the i-th agent, i=1, 2,3,4; k (k) si Suspension stiffness coefficient for the ith agent; c si Suspension damping coefficient for the ith agent; f (f) i Actuator force for the ith agent; x is x si Is the displacement at the ith agent tire;
vertical motion of the ith agent:
wherein m is usi Is the mass of the ith agent; x is x usi Is the vertical displacement of the ith agent; f (F) zi Representing the total vertical load of the ith agent; m is m si Representing the suspended mass of the ith agent; θ r Representing the road surface longitudinal gradient angle; f (F) zi Is the vertical load on the ith agent tire;
(42) State space equation of the agent: based on the dynamics model established in the step (41), the state space equation of the ith agent is obtained as follows:
in the method, in the process of the invention,u i ＝[f i ]，y i ＝[F si ]respectively the ith agentState variables, control input variables, system output variables; a is that i ，B i ，C i ，D i ，E i Respectively representing corresponding coefficient matrixes, and solving according to the dynamics model established in the step (41); d, d i Indicating the road surface stimulus to which the i-th agent is subjected.
6. The multi-agent-based active suspension output consistency control method according to claim 5, wherein the step (5) of designing the output consistency controller comprises the following specific steps:
the communication topological property among the intelligent agents is represented by graph theory knowledge, and the active suspension system is an undirected graph;
undirected graph g= { R, E } is composed of 4 vertex sets R 4 = {1,2,3,4} and edge set E; the weighted adjacency matrix of undirected graph G is denoted as a= [ a ] mn ]When edge (m, n) ∈E, a mn If the value is more than 0, the node m and the node n are called neighbor nodes, and the state information of each other is obtained; the edge set e= { (1, 2), (1, 3), (1, 4), (2, 1), (2, 3), (2, 4), (3, 1), (3, 2), (3, 4), (4, 1), (4, 2), (4, 3) };
(51) The design consistency control strategy is as follows:
in consideration of the requirement of output consistency among suspension subsystems, an active suspension system output consistency control strategy is designed based on a multi-agent theory, and the final output of an agent i consists of two parts, namely:
f i_ref ＝f i_refd +f i_refm (24)
wherein f i_ref The total output of the output consistency controller of the intelligent agent i; f (f) i_refd The method comprises the steps that the output of a force tracking controller of an intelligent agent i is obtained through solving of the force tracking controller; f (f) i_refm The method comprises the steps that the output of a force synchronous controller of an intelligent agent i is obtained through solving of the force synchronous controller;
calculating the average value of the output forces of all the agents, taking the difference between the output force of each agent and the average value to obtain a deviation value, and taking the deviation value of each agent as the input of the force synchronous controller; normalizing the output force of the intelligent agent, and calculating by adopting the normalized output force after the normalization;
the average value calculation formula is as follows:
Wherein F is m The average value of the output force of the active suspension system is obtained; f (F) i Representing the actual output force of the ith agent; f (F) i_ref Representing the expected output force of the ith agent;
the calculation formula of the deviation value of each agent is as follows:
in the formula e i_m Representing the deviation value of the ith agent;
defining the output consistency error of the intelligent agent i to embody the output consistency error among the intelligent agents, wherein the method comprises the following steps of:
wherein a is ij ∈A，H i Is a weight coefficient matrix;
the multi-agent system consistency control strategy is designed as follows:
in the formula e i_d ＝F i_ref -F i Representing a force tracking error of the ith agent; I.I.I. represents a matrix or the 2 norms of the vectors;
the output consistency of the multi-agent system can be ensured by solving the expected output of the agents meeting the strategy;
(52) Designing a force tracking controller:
solving the vehicle attitude controller designed in the step (3) to obtain the expected output force of the active suspension system, wherein the expected output force is F respectively s1 ,F s2 ,F s3 ,F s4 Taking the expected output force of the suspension subsystem as the expected output force of the corresponding agent:
F i_ref ＝F si (29)
the design control cost function is as follows:
wherein J is i_d Representing a force tracking controller control cost function value; p (P) hid Representing a force tracking controller prediction horizon; c (C) hid Representing a force tracking controller control time domain; q (Q) id1 ,Q id2 ,Q id3 Respectively representing state quantity weight matrixes of the intelligent agents i; r is R id Representing the control of the output weight matrix by the force tracking controller;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical tracking controller i_refd ；
(53) Designing a force synchronization controller:
according to the consistency control strategy designed in the step (51), the design force synchronous controller controls the cost function as follows:
wherein J is i_m Representing the value of the cost function controlled by the force synchronization controller; p (P) him Representing a predicted time domain of the force synchronization controller; c (C) him Representing the force synchronization controller control time domain; q (Q) im1 ,Q im2 ,Q im3 ,Q im4 Respectively representing related state quantity weight matrixes; r is R im The representation force synchronous controller controls the output weight matrix;
combining the control cost function and the intelligent body state space equation established in the step (42), obtaining the required control output by solving the following optimal problem:
converting the optimization problem into a quadratic programming problem, and solving the quadratic programming problem through an effective set method or an interior point method to obtain the output f of the ith intelligent physical force synchronous controller i_refm 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311199426.XA CN117111475A (en) | 2023-09-18 | 2023-09-18 | Multi-agent-based active suspension output consistency control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311199426.XA CN117111475A (en) | 2023-09-18 | 2023-09-18 | Multi-agent-based active suspension output consistency control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117111475A true CN117111475A (en) | 2023-11-24 |
Family
ID=88803854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311199426.XA Pending CN117111475A (en) | 2023-09-18 | 2023-09-18 | Multi-agent-based active suspension output consistency control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117111475A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118358310A (en) * | 2024-06-20 | 2024-07-19 | 燕山大学 | Heavy-duty vehicle composite attitude control method with electrohydraulic actuator active suspension |
-
2023
- 2023-09-18 CN CN202311199426.XA patent/CN117111475A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118358310A (en) * | 2024-06-20 | 2024-07-19 | 燕山大学 | Heavy-duty vehicle composite attitude control method with electrohydraulic actuator active suspension |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112758097A (en) | State prediction and estimation method for unmanned vehicle | |
| CN110481334B (en) | Four-wheel independent drive electric vehicle robust self-adaptive fault-tolerant control method based on disturbance observation | |
| EP1540198A2 (en) | Fuzzy controller with a reduced number of sensors | |
| CN117111475A (en) | Multi-agent-based active suspension output consistency control method | |
| Pan et al. | Fault-tolerant multiplayer tracking control for autonomous vehicle via model-free adaptive dynamic programming | |
| Ahangarnejad et al. | Integrated vehicle dynamics system through coordinating active aerodynamics control, active rear steering, torque vectoring and hydraulically interconnected suspension | |
| Masouleh et al. | Optimizing the aero-suspension interactions in a formula one car | |
| CN108681257B (en) | Design method of controller of active anti-roll system | |
| JP7446434B2 (en) | Suspension control device and suspension device control method | |
| CN112829766A (en) | Adaptive path tracking method based on distributed driving electric vehicle | |
| CN114572191A (en) | Independently-driven electric automobile trajectory tracking and stability integrated control method | |
| CN117492453B (en) | Unmodeled compensation control method for batch operation vehicles | |
| CN112487553B (en) | Design method of time lag compensation controller for controllable suspension system | |
| CN110443003B (en) | Control and optimal design method of active stabilizer bar system | |
| CN117270386A (en) | Coupling active disturbance rejection-based distributed drive six-wheel steering vehicle same-phase steering control method and controller | |
| CN116278570A (en) | Semi-automatic suspension self-adaptive cooperative control system and method for commercial vehicle based on data driving model | |
| Wang et al. | Unsprung mass effects on electric vehicle dynamics based on coordinated control scheme | |
| CN113071282B (en) | Vehicle adjusting method and device | |
| Gerard et al. | Global and local chassis control based on load sensing | |
| CN113815602B (en) | Vehicle transverse self-adaptive control method | |
| CN118119538A (en) | System and method for controlling movement of a vehicle | |
| Biglarbegian et al. | Design of a novel fuzzy controller to enhance stability of vehicles | |
| Gáspár et al. | An integrated vehicle control with actuator reconfiguration | |
| CN117841591B (en) | ISD suspension control method based on improved fuzzy neural network PID | |
| CN118358310B (en) | Heavy-duty vehicle composite attitude control method with electrohydraulic actuator active suspension |
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 |