CN109606379A - A kind of distributed driving automatic driving vehicle path trace fault tolerant control method - Google Patents
A kind of distributed driving automatic driving vehicle path trace fault tolerant control method Download PDFInfo
- Publication number
- CN109606379A CN109606379A CN201811397354.9A CN201811397354A CN109606379A CN 109606379 A CN109606379 A CN 109606379A CN 201811397354 A CN201811397354 A CN 201811397354A CN 109606379 A CN109606379 A CN 109606379A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- formula
- path trace
- follows
- tolerant 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.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
- B60W2050/0031—Mathematical model of the vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The present invention proposes a kind of distributed driving unmanned vehicle path trace fault tolerant control method, includes the following steps: distributed driving unmanned vehicle modeling；Vehicle state estimation；Unknown failure input estimation；Vehicle path trace faults-tolerant control；Tire force optimization distribution.Mathematical modeling has been carried out to distribution driving unmanned vehicle two degrees of freedom kinetic model, path trace model and unknown failure input first in the present invention.Then vehicle state estimation method and unknown failure input estimation method are devised, to provide information input for distribution driving unmanned vehicle path trace faults-tolerant control, next vehicle path trace fault tolerant control method is devised, vehicle route tracking and vehicle stabilization control target can be realized in actuator failures, finally devise tire force optimizing distribution method, for distributing the longitudinal tire force of four tires, the actuator failures problem being likely to occur in distribution driving automatic driving vehicle path tracking procedure in real time.
Description
Technical field
The invention belongs to unmanned automotive research fields, and in particular to a kind of distributed driving automatic driving vehicle path trace
Fault tolerant control method.
Background technique
As the increasingly mature and driver of control technology constantly mentions to what safety, mobility and riding comfort required
The Study of intelligent of height, vehicle has received widespread attention.The hub motor of distributed-driving electric automobile has accurately and fast
Torque response ability, this advantage for vehicle dynamic control provide more freedom degrees, be conducive to Unmanned Ground Vehicle
Intelligent traveling and stability control.The ultimate developing goal of vehicle be realize it is complete unmanned, path with
During track, the actuator of vehicle is possible to the failure for occurring unknown, once actuator failures or failure, this unknown failure
Most likely result in the reduction that serious deviation and vehicle self stability occurs in path trace, the control of entire control system
Performance, which may be decreased, even to fail, or even can inspire serious road safety issues.For example, if the steering system of vehicle occurs
Failure, will affect the divertical motion of vehicle, or even vehicle is made to lose steering capability.At this point, vehicle cannot be provided according to controller
Instruction carry out handling maneuver and track ideal path.Highly intelligentized vehicle should have processing burst actuator
The ability of failure, and ensure the validity of path following control and the stability of vehicle simultaneously when an error occurs.Therefore, it grinds
It is very necessary and significant for studying carefully unmanned vehicle path trace fault tolerant control method.
Summary of the invention
The present invention is directed to solve one of above-mentioned technical problem at least to a certain extent.
For this purpose, the present invention proposes a kind of distributed driving automatic driving vehicle path trace fault tolerant control method, the present invention
For the actuator failures problem being likely to occur in distributed driving automatic driving vehicle path tracking procedure, a kind of point is proposed
Cloth drives unmanned vehicle path trace fault tolerant control method.Unmanned vehicle two degrees of freedom power is driven to distribution first in the present invention
It learns model, path trace model and unknown failure input and has carried out mathematical modeling.Then vehicle state estimation method is devised
Estimation method is inputted with unknown failure, to provide information input for distribution driving unmanned vehicle path trace faults-tolerant control, is connect
Get off to devise vehicle path trace fault tolerant control method, can realize that vehicle route tracking is steady with vehicle in actuator failures
Qualitative contrlol target finally devises tire force optimizing distribution method, for distributing the longitudinal tire force of four tires in real time.
The technical scheme is that
A kind of distributed driving automatic driving vehicle path trace fault tolerant control method, comprising the following steps:
Distribution driving unmanned vehicle modeling: it is modeled including the modeling of two degrees of freedom dynamics of vehicle, path trace and unknown
Failure input modeling；
Vehicle state estimation: using the method design vehicle state observer of sliding mode observer, estimation obtains vehicle-state；
Unknown failure input estimation: according to resulting vehicle-state is estimated, unknown failure input is estimated；
Vehicle path trace faults-tolerant control: the unknown failure obtained according to estimation inputs the input quantity as fault-tolerant controller
Obtain vehicle path trace faults-tolerant control rate；
Tire force optimization distribution: the target letter of tire force optimization distribution is obtained according to vehicle path trace faults-tolerant control rate
Number carries out tire force optimization distribution.
In above scheme, the lateral direction of car and sideway kinetics equation of the two degrees of freedom dynamics of vehicle modeling distinguish table
It is shown as:
In formula, vxFor longitudinal speed, vyFor lateral speed, γ is yaw velocity, and m is car mass, IzFor around z-axis
Rotary inertia, FyfAnd FyrRespectively front and back lateral tire forces, lfDistance for mass center away from front axle, lrFor mass center away from rear axle away from
From MzFor the vehicle yaw moment as caused by four longitudinal force of tire:
In formula, δfFor front wheel angle, bsFor half wheelspan of antero posterior axis, FxiFor the longitudinal force of i-th of tire, lateral tire forces
It indicates are as follows:
In formula, CfFor front tyre cornering stiffness, CrFor rear tyre cornering stiffness, αfFor front tyre side drift angle, αrFor rear tyre
Side drift angle, slip angle of tire indicate are as follows:
It is obtained by formula one and formula two:
Wherein β is vehicle centroid side drift angle, and meets β=arctan (vy/vx)。
In above scheme, the path trace modeling, using the lateral deviation and boat between actual vehicle and ideal path
Path trace effect is characterized to deviation, the course deviation in vehicle route trace model and its differential equation established indicate
Are as follows:
In formula, ψ is course deviation, ψhFor actual vehicle course angle, ψdFor desired vehicle course angle；ψhAnd ψdIt indicates
Are as follows:
In formula, ρ is the radius of curvature of expected path；
Path trace lateral deviation equation are as follows:
Formula ten simplifies are as follows:
In above scheme, the unknown failure input modeling specifically:
The state space equation of path trace faults-tolerant control indicates are as follows:
In formula, state vector is x=[β γ ψ e]T, measurement input is y=[γ ψ e]T, control input is u=[δf
Mz]T, input matrix isCalculation matrix is C=[03×1 I3×3], state transition function is f (x, t)=[f1 f2]T,
Wherein the two subsystems of state transition function are respectivelyG (t) is that unknown failure is defeated
Enter, enables
In above scheme, the vehicle state estimation specifically:
The vehicle-state observer indicates are as follows:
In formula, H is observer gain, and λ is State observer switching item；
It is described to design State observer switching item are as follows:
In formula,Represent output error, ξ1The thickness in vehicle-state observer boundary layer is represented, μ represents switching item
Gain matrix；
By the alternative manner of turn off gain matrix, according to the dynamic change that unknown failure inputs, the switch of adjustment in real time increases
Beneficial matrix, the alternative manner are as follows:
μi,k+1(t)=μi,k(t)+Ki||eyi,k(t)|-ξ0|×sgn{(|eyi,k(t)|-ξ0)×(|eyi,k-1(t)|-ξ0)}
Formula 15
Wherein i=1,2,3, μ=[μ1 μ2 μ3]T, μi,k+1It (t) is+1 value of μ kth, μi,kIt (t) is k-th of value of μ,
KiFor corresponding μiGain matrix, and it acts as adjustment iterative algorithm convergence rate；μ in vector μ1, μ2, μ3It is right respectively
Answer yaw rate, course deviation and lateral deviation.
In above scheme, the estimation equation of the unknown failure input in the unknown failure input estimation is indicated:
In formula, T is time step.
In above scheme, the vehicle path trace faults-tolerant control specifically:
As the input quantity of fault-tolerant controller, to obtain vehicle path trace fault-tolerant for the unknown failure input obtained according to estimation
Control rate；
Path trace faults-tolerant control state space equation in decoupling formula 12 can obtain:
In formula, x1=[β γ]T, x2=[ψ e]T, in vehicle route trace model, entire vehicle regards a movement as
Point, so as toB2=0, g2(t)=0；
Sliding-mode surface s are as follows:
S=(x1-x1r)+k(x2-x2r) formula 18
K is control matrix, x in formula1r、x2rIt respectively represents and x1、x2Corresponding ideal vehicle-state, and x1r=[0
γd]T, x2r=0, γdFor ideal yaw velocity；
The control rate of vehicle path trace fault-tolerant controller are as follows:
In formula, u1For conventional control item, u2For switching control item；
u1It indicates are as follows:
u2It indicates are as follows:
In formula, ξ2Represent the thickness in vehicle path trace fault-tolerant controller boundary layer, τ vehicle path trace fault-tolerant controller
Switching control gain andε is the constant greater than 0.
In above scheme, the objective function of the tire force optimization distribution is defined as:
In formula, J1For objective function longitudinal direction assignment item, J2For objective function transverse distribution item, Fx=[Fx1 Fx2 Fx3 Fx4]T
For longitudinal force of tire, matrixW=diag [w1 w2 w3 w4] it is control point
With matrix, kf(0 < kf< 1) it is weight coefficient；ByObtain the optimal solution of objective function in formula 22 are as follows:
Control distribution allocation matrix W=diag [w1 w2 w3 w4] in parameter value be
In above scheme, unknown failure is inputted and carries out grade evaluation, chosen same when estimating with unknown failure input
Time step T calculates the mean value g of unknown failure inputmWith mean square deviation gvFoundation as grade evaluation；
If data point is n, forI.e.When, average calculation method is
It averages to all data points since 0 to current time, asCorresponding mean square deviation calculating side
Method is
WhenWhen, average calculation method be to since current time data point to before the data pointIt is a
Data point is averaged, asMean square deviation calculation method is accordingly
In above scheme, the threshold value of unknown failure input rank evaluation is chosen；
As the mean value g of unknown failure input estimated resultmWhen less than threshold value, weight coefficient value at this time are as follows:
In formula, kf0For a constant, so that kfBetween 0 to 1；For the absolute value of slip angle estimation result, | | gm|
| for the norm of unknown failure input estimation mean value；
As the mean value g of unknown failure input estimated resultmWhen greater than threshold value, weight coefficient value at this time are as follows:
In formula, | | gv| | for the norm of unknown failure input estimation mean square deviation.
Compared with prior art, the beneficial effects of the present invention are:
1, unmanned vehicle can develop towards height autonomy-oriented, intelligentized direction, in path tracking procedure, in fact it could happen that
Actuator failures influence whether the normal operation of unmanned vehicle control.The present invention considers this factor, proposes a kind of point
Cloth drives automatic driving vehicle path trace fault tolerant control method, drives unmanned vehicle two degrees of freedom dynamics to distribution first
Model, path trace model and unknown failure input have carried out mathematical modeling.Then devise vehicle state estimation method and
Unknown failure inputs estimation method, to provide information input for distribution driving unmanned vehicle path trace faults-tolerant control, connects down
Vehicle path trace fault tolerant control method is devised, vehicle route tracking and vehicle stabilization can be realized in actuator failures
Property control target, finally devise tire force optimizing distribution method, for distributing the longitudinal tire force of four tires in real time, we
Method can help unmanned vehicle in actuator failures, realizing route tracing control, while guarantee the Yaw stability of vehicle.
2, unmanned vehicle part biography can be substituted based on Design of Sliding Mode Observer unmanned vehicle method for estimating state in the present invention
The function of sensor reduces unmanned vehicle cost.
3, failure has been carried out using the estimated result that unknown failure inputs and has been commented when tire force optimizes and distributes in the present invention
Valence facilitates the adaptive ability of further Lifting Control System, makes it that can better adapt to and handle in practical applications reality
The failure problems being likely to occur in the driving cycle of border.
4, the present invention devises the grade evaluation mechanism of unknown failure, and the setting of failure mean square deviation helps to improve failure and estimates
The reliability and stability of meter, avoiding individual data point fluctuation causes to control not normal phenomenon；The setting of fault threshold helps
In improving weight system to the sensibility of failure, to guarantee the effect of tire force optimization distribution.
Detailed description of the invention
Above-mentioned and/or additional aspect of the invention and advantage will become from the description of the embodiment in conjunction with the following figures
Obviously and it is readily appreciated that, in which:
Fig. 1 is distributed driving unmanned vehicle path trace fault tolerant control method flow chart.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
The orientation or positional relationship of the instructions such as " thickness ", "upper", "lower", " axial direction ", " radial direction ", "vertical", "horizontal", "inner", "outside"
To be based on the orientation or positional relationship shown in the drawings, be merely for convenience of description of the present invention and simplification of the description, rather than indicate or
It implies that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore should not be understood as
Limitation of the present invention.In addition, term " first ", " second " are used for description purposes only, it is not understood to indicate or imply phase
To importance or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be with
Explicitly or implicitly include one or more of the features.In the description of the present invention, the meaning of " plurality " is two or
Two or more, unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected；It can be machine
Tool connection, is also possible to be electrically connected；It can be directly connected, two members can also be can be indirectly connected through an intermediary
Connection inside part.For the ordinary skill in the art, above-mentioned term can be understood in this hair as the case may be
Concrete meaning in bright.
Distributed driving unmanned vehicle path trace according to an embodiment of the present invention is specifically described in conjunction with attached drawing first below to hold
Wrong control method.
Fig. 1 show a kind of reality of distributed driving automatic driving vehicle path trace fault tolerant control method of the present invention
Mode is applied, the distributed driving automatic driving vehicle path trace fault tolerant control method includes the following steps:
Distribution driving unmanned vehicle modeling: it is modeled including the modeling of two degrees of freedom dynamics of vehicle, path trace and unknown
Failure input modeling.
The modeling of two degrees of freedom dynamics of vehicle.By the origin of dynamic coordinate system xyz be fixed on vehicle and and vehicle's center of gravity
It is overlapped, x axis is overlapped with the longitudinal driving direction of vehicle, and y-axis is overlapped with the transverse movement direction of vehicle, the longitudinal direction of z-axis and vehicle
Driving direction is overlapped.Lateral direction of car and sideway kinetics equation can respectively indicate are as follows:
In formula, vxFor longitudinal speed, vyFor lateral speed, γ is yaw velocity, and m is car mass, IzFor around z-axis
Rotary inertia, FyfAnd FyrRespectively front and back lateral tire forces, lfDistance for mass center away from front axle, lrFor mass center away from rear axle away from
From MzFor the vehicle yaw moment as caused by four longitudinal force of tire
In formula, δfFor front wheel angle, bsFor half wheelspan of antero posterior axis, FxiIt (i=1,2,3,4) is the longitudinal direction of i-th of tire
Power.Lateral tire forces may be expressed as:
In formula, CfAnd CrRespectively front and back tire cornering stiffness, αfAnd αrFor front and back slip angle of tire.Slip angle of tire can
It is expressed as
Then it can be obtained by formula one and formula two:
Path trace modeling.In vehicle route trace model using between actual vehicle and ideal path lateral deviation with
Course deviation characterizes path trace effect, and the course deviation in vehicle route trace model and its differential equation established can
It is expressed as
In formula, ψ is course deviation, ψhFor actual vehicle course angle, ψdFor desired vehicle course angle.ψhAnd ψdIt can table
It is shown as
In formula, ρ is the radius of curvature of expected path.Path trace lateral deviation equation are as follows:
In view of course deviation angle is comparatively smaller, then formula 12 can simplify are as follows:
Unknown failure input modeling.It enables Simultaneous formula six, formula seven, formula eight and formula ten, while in view of possible
The state space equation of the actuator failures of appearance, path trace faults-tolerant control may be expressed as:
In formula, state vector is x=[β γ ψ e]T, measurement input is y=[γ ψ e]T, control input is u=[δf
Mz]T, input matrix isCalculation matrix is C=[03×1 I3×3], state transition function is f (x, t)=[f1 f2]T,
Wherein the two subsystems of state transition function are respectivelyG (t) is that unknown failure is defeated
Enter.It include system interference, unmodeled disturbance and actuator failures in unknown failure input.Even if actuator is without failure,
System interference and unmodeled disturbance also will affect the effect of path following control device, therefore system interference and unmodeled disturbance can not
Ignore.When actuator breaks down, these three factors can influence the effect of path following control together, therefore, at this time they
It is considered as the unknown failure input of path trace faults-tolerant control state space equation together.It should be noted that in actuator event
When barrier modeling, actuator failures are not assigned to the sub- actuator of some part specifically, but entire vehicle dynamics system exists
It is considered as an actuator in formula 12.When a certain section failure of vehicle dynamics system, unknown failure input
Actually indicate that actuator failures (also including system interference and unmodeled disturbance) influence path tracking control system.In order to disappear
Except this influence, the design object of path trace fault tolerant control method can be also protected simultaneously even if when actuator breaks down
Demonstrate,prove path trace precision and intact stability.Distribution driving unmanned vehicle modeling is all estimation methods and control method below one
Basis.
The design of vehicle state estimation method.
For the vehicle route tracking system in formula 12, in order to which design path tracks fault-tolerant controller, it is necessary first to
Unknown failure input is estimated.Since the direct measurement difficulty of vehicle centroid side drift angle is larger, measurement cost is higher, and
The measured values such as the yaw velocity in quantity of state, course error, lateral shift in formula 12 are also by the shadow for measuring noise
It rings, therefore before carrying out Unknown worm Fault Estimation, devises a kind of vehicle-state observer using sliding mode observer method,
It can indicate are as follows:
In formula, H is observer gain, and λ is State observer switching item.Due to the presence of State observer switching item, observer is estimated
Inevitably there is chattering phenomenon in meter result.In order to weaken this high frequency chattering phenomenon, introduced near sliding mode
Boundary layer, and State observer switching item is designed are as follows:
In formula,Represent output error, ξ1The thickness in vehicle-state observer boundary layer is represented, μ represents switching item
Gain matrix.
In path following control, since actuator failures and system interference are unpredictable, and its upper limit value is also
Dynamic change.Therefore, when designing μ, in order to avoid the error being likely to occur, it should choose a relatively large μ.
But if the μ chosen is too big, it would be possible to lead to the disturbance of estimated result and unstable.In order to solve this problem,
A kind of alternative manner of turn off gain matrix is proposed, the dynamic change that this method is inputted according to unknown failure, adjustment is opened in real time
Gain matrix is closed, even if vehicle-state observer is also able to achieve preferable estimation performance to guarantee in Actuators Failures.This
One alternative manner are as follows:
μi,k+1(t)=μi,k(t)+Ki||eyi,k(t)|-ξ0|×sgn{(|eyi,k(t)|-ξ0)×(|eyi,k-1(t)|-ξ0)}
Formula 15
Wherein i=1,2,3, μ=[μ1 μ2 μ3]T, μi,k+1It (t) is+1 value of μ kth, μi,kIt (t) is k-th of value of μ,
KiFor corresponding μiGain matrix and it acts as adjustment iterative algorithm convergence rate.That need to show is vector μiIn
μ1, μ2, μ3Respectively correspond yaw rate, course deviation and lateral deviation.According to formula 15 it can be found that if
Vehicle-state observer is not or not sliding formwork boundary, then the gain matrix of first term will be adjusted to larger quantities, at this time symbol letter
Several calculated result is 1.When vehicle-state observer reaches sliding formwork boundary, the calculated result of sign function is -1, switch
The gain matrix of item will reduce.
Unknown failure inputs estimation method design.
When unmanned vehicle actuator breaks down, failure and system interference are considered as the unknown of vehicle route tracking system
Failure input.Using resulting vehicle-state is estimated in step S2, unknown failure input can be further estimated, is subsequently used for
The unknown failure estimated inputs the input quantity as fault-tolerant controller.Preferably, selecting time step T is 0.01 second, and will
Its iteration cycle for being set as yes vehicle route tracking system, then the estimation equation of unknown failure input can indicate:
In formula, T is time step.It should be noted that in the actual intelligent traveling of vehicle or path tracking procedure
In, either certain unpredictable components of hub motor, wheel steering system or other executing agencies have generation
A possibility that failure, therefore be difficult to judge that failure partially has occurred in which of executing agency in time.Proposed in formula 16
The estimation method of unknown failure input can assess the damage caused by control system of unknown failure and interference, without knowing
Where is specific failure generation.That is, estimated unknown failure input is not substantially the size of failure itself,
But failure gives path trace tolerant system bring interference level on a control level.This estimation thought more meets unmanned vehicle
Practical situations help to improve the scope of application of fault tolerant control method designed by the present invention.
The design of vehicle path trace fault tolerant control method.
During Vehicular intelligent traveling and path following control, the catastrophic failure of actuator will lead to intact stability
With the inaccuracy even severe deviations of path following control performance.Therefore, path following control device should be fault-tolerant or to failure
It is insensitive, it can guarantee while being influenced by actuator failures and some unknown disturbances vehicle stability and path with
Track precision.Path trace faults-tolerant control state space equation in decoupling formula 12 can obtain:
In formula, x1=[β γ]T, x2=[ψ e]T.In vehicle route trace model, it can be seen that entire vehicle is practical
On be counted as a mobile point, so as toB2=0, g2(t)=0.Sliding-mode surface s are as follows:
S=(x1-x1r)+k(x2-x2r) formula 18
K is control matrix, x in formula1r、x2rIt respectively represents and x1、x2Corresponding ideal vehicle-state, and x1r=[0
γd]T, x2r=0, γdFor ideal yaw velocity.
The control rate of vehicle path trace fault-tolerant controller is designed to:
In formula, u1For conventional control item, u2For switching control item.u1It can indicate are as follows:
In design u2During, it is contemplated that the switching possible control wild effect of item, it can be by u2Design are as follows:
In formula, ξ1The thickness in vehicle path trace fault-tolerant controller boundary layer is represented, τ represents the fault-tolerant control of vehicle path trace
The switching control gain of device processed andε is the constant greater than 0.
The design of tire force optimizing distribution method.
The objective function of tire force optimization distribution is defined as:
In formula, J1For objective function longitudinal direction assignment item, J2For objective function transverse distribution item, Fx=[Fx1 Fx2 Fx3 Fx4]T
For longitudinal force of tire, matrix BxIt can be calculated and be represented by by formula threeW=diag [w1 w2 w3 w4] it is control allocation matrix, kf(0 < kf<
It 1) is weight coefficient.ByThe optimal solution of objective function in formula 22 can be obtained are as follows:
Control distribution allocation matrix W=diag [w1 w2 w3 w4] in parameter value be
Influence of the different unknown failure input ranks to path trace fault-tolerant control system is of different sizes, so needs pair
Estimate that resulting unknown failure input carries out grade evaluation in unknown failure input estimation.Carry out the evaluation of unknown failure input rank
When, it chooses and inputs same time step T when estimating with unknown failure in unknown failure input estimation, as 0.01 second, in addition,
Calculate the mean value g of unknown failure inputmWith mean square deviation gvFoundation as grade evaluation.If data point is n, in first 1 second
100 data points, i.e. when 0≤n≤100, average calculation method is to ask flat to all data points since 0 to current time
Mean value, asMean square deviation calculation method is accordinglyAs n >=100,
Average calculation method is to average to since current time data point to 99 data points before the data point, asMean square deviation calculation method is accordinglyThis calculation method is advantageous
There is a more reliable assessment in inputting to current unknown failure, and will not cause to assess because of the exception of individual data point
As a result inaccuracy.
In the objective function in formula 22, J1The larger specific gravity for occupying longitudinal tire force distribution, for ensuring to indulge
To driveability, and J2There is direct influence on the stability control of vehicle.Therefore, if lateral stability of cars situation compared with
The grade evaluation result of difference, unknown failure input estimation is more serious, then should will reduce weight coefficient kf, to improve J2Control effect
Fruit, to improve vehicle yaw stability and path trace precision.Conversely, weight coefficient should be adjusted bigger, increase J1Institute
The control ratio accounted for, to improve the responding ability of tire force.Simultaneously, it is contemplated that the amount of yaw rate, side slip angle
Grade chooses 0.1 as the threshold value of unknown failure input rank evaluation.Preferably, the threshold values is 0.1；When unknown failure input is estimated
Count the mean value g of resultmWhen less than 0.1, it is believed that fault level is lower, at this time weight coefficient value are as follows:
In formula, kf0For a constant, so that kfBetween 0 to 1；For the absolute value of slip angle estimation result, | | gm||
The norm of estimation mean value is inputted for unknown failure.As the mean value g of unknown failure input estimated resultmWhen greater than 0.1, it is believed that failure
Higher ranked, the mean square deviation of unknown failure input also can generate large effect to the allocation result of tire force, at this time weight system
Number value are as follows:
In formula, | | gv| | for the norm of unknown failure input estimation mean square deviation.
Although not each embodiment only includes one it should be appreciated that this specification describes according to various embodiments
A independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should will say
As a whole, the technical solutions in the various embodiments may also be suitably combined for bright book, and forming those skilled in the art can be with
The other embodiments of understanding.
The series of detailed descriptions listed above are illustrated only for possible embodiments of the invention,
The protection scope that they are not intended to limit the invention, it is all without departing from equivalent embodiment made by technical spirit of the present invention or change
It should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of distributed driving automatic driving vehicle path trace fault tolerant control method, which comprises the following steps:
Distribution driving unmanned vehicle modeling: including the modeling of two degrees of freedom dynamics of vehicle, path trace modeling and unknown failure
Input modeling；
Vehicle state estimation: using the method design vehicle state observer of sliding mode observer, estimation obtains vehicle-state；
Unknown failure input estimation: according to resulting vehicle-state is estimated, unknown failure input is estimated；
Vehicle path trace faults-tolerant control: the unknown failure input obtained according to estimation is obtained as the input quantity of fault-tolerant controller
Vehicle path trace faults-tolerant control rate；
Tire force optimization distribution: obtaining the objective function of tire force optimization distribution according to vehicle path trace faults-tolerant control rate, into
The optimization distribution of row tire force.
2. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 1, feature exist
In,
The lateral direction of car and sideway kinetics equation of the two degrees of freedom dynamics of vehicle modeling respectively indicate are as follows:
In formula, vxFor longitudinal speed, vyFor lateral speed, γ is yaw velocity, and m is car mass, IzFor around the rotation of z-axis
Inertia, FyfAnd FyrRespectively front and back lateral tire forces, lfDistance for mass center away from front axle, lrDistance for mass center away from rear axle, Mz
For the vehicle yaw moment as caused by four longitudinal force of tire:
In formula, δfFor front wheel angle, bsFor half wheelspan of antero posterior axis, FxiFor the longitudinal force of i-th of tire, lateral tire forces are indicated
Are as follows:
In formula, CfFor front tyre cornering stiffness, CrFor rear tyre cornering stiffness, αfFor front tyre side drift angle, αrFor rear tyre lateral deviation
Angle, slip angle of tire indicate are as follows:
It is obtained by formula one and formula two:
Wherein β is vehicle centroid side drift angle, and meets β=arctan (vy/vx)。
3. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 2, feature exist
In the path trace modeling characterizes path with course deviation using the lateral deviation between actual vehicle and ideal path
Tracking effect, the course deviation in vehicle route trace model and its differential equation established indicate are as follows:
In formula, ψ is course deviation, ψhFor actual vehicle course angle, ψdFor desired vehicle course angle；ψhAnd ψdIt indicates are as follows:
In formula, ρ is the radius of curvature of expected path；
Path trace lateral deviation equation are as follows:
Formula ten simplifies are as follows:
4. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 3, feature exist
In the unknown failure input modeling specifically:
The state space equation of path trace faults-tolerant control indicates are as follows:
In formula, state vector is x=[β γ ψ e]T, measurement input is y=[γ ψ e]T, control input is u=[δf Mz]T,
Input matrix isCalculation matrix is C=[03×1 I3×3], state transition function is f (x, t)=[f1 f2]T,
The two subsystems of middle state transition function are respectivelyG (t) is that unknown failure is defeated
Enter, enables
5. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 4, feature exist
In the vehicle state estimation specifically:
The vehicle-state observer indicates are as follows:
In formula, H is observer gain, and λ is State observer switching item；
It is described to design State observer switching item are as follows:
In formula,Represent output error, ξ1The thickness in vehicle-state observer boundary layer is represented, μ represents the increasing of switching item
Beneficial matrix；
Turn off gain square is adjusted in real time according to the dynamic change that unknown failure inputs by the alternative manner of turn off gain matrix
Battle array, the alternative manner are as follows:
μi,k+1(t)=μi,k(t)+Ki||eyi,k(t)|-ξ0|×sgn{(|eyi,k(t)|-ξ0)×(|eyi,k-1(t)|-ξ0)}
Formula 15
Wherein i=1,2,3, μ=[μ1 μ2 μ3]T, μi,k+1It (t) is+1 value of μ kth, μi,kIt (t) is k-th of value of μ, KiIt is right
The μ answerediGain matrix, and it acts as adjustment iterative algorithm convergence rate；μ in vector μ1, μ2, μ3It respectively corresponds
Yaw rate, course deviation and lateral deviation.
6. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 5, feature exist
In the estimation equation of the unknown failure input in the unknown failure input estimation indicates:
In formula, T is time step.
7. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 6, feature exist
In the vehicle path trace faults-tolerant control specifically:
The unknown failure input obtained according to estimation obtains vehicle path trace faults-tolerant control as the input quantity of fault-tolerant controller
Rate；
Path trace faults-tolerant control state space equation in decoupling formula 12 can obtain:
In formula, x1=[β γ]T, x2=[ψ e]T, in vehicle route trace model, entire vehicle regards a mobile point as,
So as toB2=0, g2(t)=0；
Sliding-mode surface s are as follows:
S=(x1-x1r)+k(x2-x2r) formula 18
K is control matrix, x in formula1r、x2rIt respectively represents and x1、x2Corresponding ideal vehicle-state, and x1r=[0 γd]T,
x2r=0, γdFor ideal yaw velocity；
The control rate of vehicle path trace fault-tolerant controller are as follows:
In formula, u1For conventional control item, u2For switching control item；
u1It indicates are as follows:
u2It indicates are as follows:
In formula, ξ2The thickness in vehicle path trace fault-tolerant controller boundary layer is represented, τ vehicle path trace fault-tolerant controller is cut
Change control gain andε is the constant greater than 0.
8. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 7, feature exist
In the objective function of the tire force optimization distribution is defined as:
In formula, J1For objective function longitudinal direction assignment item, J2For objective function transverse distribution item, Fx=[Fx1 Fx2 Fx3 Fx4]TFor wheel
Tire longitudinal force, matrixW=diag [w1 w2 w3 w4] it is control distribution moments
Battle array, kf(0 < kf< 1) it is weight coefficient；ByObtain the optimal solution of objective function in formula 22 are as follows:
Control distribution allocation matrix W=diag [w1 w2 w3 w4] in parameter value be
9. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 8, feature exist
In to unknown failure input progress grade evaluation, same time step T when choosing with unknown failure input estimation, calculating is not
Know the mean value g of failure inputmWith mean square deviation gvFoundation as grade evaluation；
If data point is n,I.e.When, average calculation method is to from 0
Start to all data points at current time to average, asMean square deviation calculation method is accordingly
WhenWhen, average calculation method be to since current time data point to before the data pointA data point
It averages, asMean square deviation calculation method is accordingly
10. distributed driving automatic driving vehicle path trace fault tolerant control method according to claim 9, feature exist
In the threshold value of selection unknown failure input rank evaluation；
As the mean value g of unknown failure input estimated resultmWhen less than threshold value, weight coefficient value at this time are as follows:
In formula, kf0For a constant, so that kfBetween 0 to 1；For the absolute value of slip angle estimation result, | | gm| | for not
Know the norm of failure input estimation mean value；
As the mean value g of unknown failure input estimated resultmWhen greater than threshold value, weight coefficient value at this time are as follows:
In formula, | | gv| | for the norm of unknown failure input estimation mean square deviation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811397354.9A CN109606379B (en) | 2018-11-22 | 2018-11-22 | Path tracking fault-tolerant control method for distributed driving unmanned vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811397354.9A CN109606379B (en) | 2018-11-22 | 2018-11-22 | Path tracking fault-tolerant control method for distributed driving unmanned vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109606379A true CN109606379A (en) | 2019-04-12 |
| CN109606379B CN109606379B (en) | 2020-07-31 |
Family
ID=66004388
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811397354.9A Active CN109606379B (en) | 2018-11-22 | 2018-11-22 | Path tracking fault-tolerant control method for distributed driving unmanned vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109606379B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110758404A (en) * | 2019-11-28 | 2020-02-07 | 吉林大学 | Fault-tolerant control method for actuator fault of vehicle stabilization system |
| CN110782092A (en) * | 2019-10-25 | 2020-02-11 | 北京京东乾石科技有限公司 | Trajectory planning method and device of unmanned distribution vehicle in unstructured scene |
| CN111007722A (en) * | 2019-12-18 | 2020-04-14 | 厦门大学 | Transverse robust fault-tolerant control system and method for four-wheel steering automatic driving automobile |
| CN111428382A (en) * | 2020-04-17 | 2020-07-17 | 福瑞泰克智能系统有限公司 | Method, system, computer device and readable storage medium for vehicle trajectory control |
| CN111959527A (en) * | 2020-08-07 | 2020-11-20 | 长春工业大学 | Automobile path tracking control method based on corner optimization sequence |
| CN112051851A (en) * | 2020-09-09 | 2020-12-08 | 清华大学 | Autonomous drift control method and system for electric four-wheel drive vehicle under limit working condition |
| CN112102616A (en) * | 2020-09-09 | 2020-12-18 | 常熟理工学院 | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model |
| CN112550294A (en) * | 2020-11-16 | 2021-03-26 | 东南大学 | Path tracking control method based on vehicle fault signal isolation |
| CN112606826A (en) * | 2020-12-24 | 2021-04-06 | 江苏大学 | State variable full-constraint direct yaw moment control algorithm based on road adhesion coefficient |
| CN113126623A (en) * | 2021-04-13 | 2021-07-16 | 吉林大学 | Adaptive dynamic sliding mode automatic driving vehicle path tracking control method considering input saturation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007281779A (en) * | 2006-04-05 | 2007-10-25 | Sumitomo Electric Ind Ltd | Method for fault detection and control of optical line |
| CN103488080A (en) * | 2013-09-09 | 2014-01-01 | 河北科技师范学院 | Lunar rover coordinated drive self-adaption fault-tolerant control method based on hierarchical fuzzy system |
| CN104401325A (en) * | 2014-11-05 | 2015-03-11 | 江苏大学 | Dynamic regulation and fault tolerance method and dynamic regulation and fault tolerance system for auxiliary parking path |
| CN106125550A (en) * | 2016-07-20 | 2016-11-16 | 南京航空航天大学 | The combined failure of a kind of high ferro traction rectifier device is estimated and fault tolerant control method |
| CN108052106A (en) * | 2018-01-19 | 2018-05-18 | 东南大学 | A kind of unmanned agricultural machinery path following system based on ROS |
| CN108454623A (en) * | 2018-01-22 | 2018-08-28 | 大连理工大学 | A kind of unmanned electric vehicle Trajectory Tracking Control method of four motorized wheels |
-
2018
- 2018-11-22 CN CN201811397354.9A patent/CN109606379B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007281779A (en) * | 2006-04-05 | 2007-10-25 | Sumitomo Electric Ind Ltd | Method for fault detection and control of optical line |
| CN103488080A (en) * | 2013-09-09 | 2014-01-01 | 河北科技师范学院 | Lunar rover coordinated drive self-adaption fault-tolerant control method based on hierarchical fuzzy system |
| CN104401325A (en) * | 2014-11-05 | 2015-03-11 | 江苏大学 | Dynamic regulation and fault tolerance method and dynamic regulation and fault tolerance system for auxiliary parking path |
| CN106125550A (en) * | 2016-07-20 | 2016-11-16 | 南京航空航天大学 | The combined failure of a kind of high ferro traction rectifier device is estimated and fault tolerant control method |
| CN108052106A (en) * | 2018-01-19 | 2018-05-18 | 东南大学 | A kind of unmanned agricultural machinery path following system based on ROS |
| CN108454623A (en) * | 2018-01-22 | 2018-08-28 | 大连理工大学 | A kind of unmanned electric vehicle Trajectory Tracking Control method of four motorized wheels |
Non-Patent Citations (1)
| Title |
|---|
| RONGRONGWANG, CHUAN HU: "Composite Nonlinear Feedback Control for Path Following of Four-Wheel Independently Actuated Autonomous Ground Vehicles", 《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110782092A (en) * | 2019-10-25 | 2020-02-11 | 北京京东乾石科技有限公司 | Trajectory planning method and device of unmanned distribution vehicle in unstructured scene |
| CN110758404A (en) * | 2019-11-28 | 2020-02-07 | 吉林大学 | Fault-tolerant control method for actuator fault of vehicle stabilization system |
| CN111007722A (en) * | 2019-12-18 | 2020-04-14 | 厦门大学 | Transverse robust fault-tolerant control system and method for four-wheel steering automatic driving automobile |
| CN111428382A (en) * | 2020-04-17 | 2020-07-17 | 福瑞泰克智能系统有限公司 | Method, system, computer device and readable storage medium for vehicle trajectory control |
| CN111959527A (en) * | 2020-08-07 | 2020-11-20 | 长春工业大学 | Automobile path tracking control method based on corner optimization sequence |
| CN111959527B (en) * | 2020-08-07 | 2023-09-05 | 长春工业大学 | Automobile path tracking control method based on corner optimization sequence |
| CN112102616A (en) * | 2020-09-09 | 2020-12-18 | 常熟理工学院 | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model |
| CN112051851B (en) * | 2020-09-09 | 2022-03-01 | 清华大学 | Autonomous drift control method and system for electric four-wheel drive vehicle under limit working condition |
| CN112051851A (en) * | 2020-09-09 | 2020-12-08 | 清华大学 | Autonomous drift control method and system for electric four-wheel drive vehicle under limit working condition |
| CN112550294A (en) * | 2020-11-16 | 2021-03-26 | 东南大学 | Path tracking control method based on vehicle fault signal isolation |
| CN112606826A (en) * | 2020-12-24 | 2021-04-06 | 江苏大学 | State variable full-constraint direct yaw moment control algorithm based on road adhesion coefficient |
| CN112606826B (en) * | 2020-12-24 | 2022-02-15 | 江苏大学 | State variable full-constraint direct yaw moment control algorithm based on road adhesion coefficient |
| CN113126623A (en) * | 2021-04-13 | 2021-07-16 | 吉林大学 | Adaptive dynamic sliding mode automatic driving vehicle path tracking control method considering input saturation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109606379B (en) | 2020-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109606379A (en) | A kind of distributed driving automatic driving vehicle path trace fault tolerant control method | |
| CN108594652B (en) | Observer information iteration-based vehicle state fusion estimation method | |
| CN103057436B (en) | Yawing moment control method of individual driven electromobile based on multi-agent | |
| CN103158703B (en) | A kind of control method of Parallel parking | |
| US9950703B2 (en) | Vehicle with independently driven multiple axes, and controller which independently drives multiple axles | |
| Soudbakhsh et al. | A collision avoidance steering controller using linear quadratic regulator | |
| CN111007722B (en) | Transverse robust fault-tolerant control system and method for four-wheel steering automatic driving automobile | |
| CN109733205A (en) | A kind of wheel hub electric car direct yaw moment control method with fault tolerance | |
| CN105416276A (en) | Method for controlling electric automobile stability direct yawing moment based on high-order slip mold | |
| Lian et al. | Cornering stiffness and sideslip angle estimation based on simplified lateral dynamic models for four-in-wheel-motor-driven electric vehicles with lateral tire force information | |
| CN106649983A (en) | Vehicle dynamics model modeling method used in unmanned vehicle high-velocity motion planning | |
| Jeon et al. | Development of an electric active rollcontrol (ARC) algorithm for a SUV | |
| CN104477237A (en) | Four wheel independent steering electric car steering control method and system | |
| CN106696760A (en) | Power distribution method for hub-motor-driven vehicle | |
| CN108216250A (en) | Four-drive electric car speed and road grade method of estimation based on state observer | |
| CN108597058A (en) | Distributed-driving electric automobile state based on pseudo- measurement information cascades method of estimation | |
| CN110481334A (en) | Four motorized wheels electric car robust adaptive fault tolerant control method based on disturbance observation | |
| CN112034706B (en) | Mobile robot fault-tolerant control method and equipment based on multi-mode switching | |
| CN110979026A (en) | Distributed driving bus torque distribution method based on real-time road conditions | |
| Huang et al. | Lateral stability control of four-wheel independent drive electric vehicles based on model predictive control | |
| Kang et al. | Design and testing of a controller for autonomous vehicle path tracking using GPS/INS sensors | |
| CN106168758B (en) | course tracking control method of four-wheel independent drive electric vehicle | |
| Stolte et al. | Toward Fault-Tolerant Vehicle Motion Control for Over-Actuated Automated Vehicles: A Non-Linear Model Predictive Approach | |
| Chen et al. | Reliable sideslip angle estimation of four-wheel independent drive electric vehicle by information iteration and fusion | |
| Li et al. | AFS/DYC control of in-wheel motor drive electric vehicle with adaptive tire cornering stiffness |
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 |