CN109969181A - A kind of deviation auxiliary system and its deviation householder method - Google Patents
A kind of deviation auxiliary system and its deviation householder method Download PDFInfo
- Publication number
- CN109969181A CN109969181A CN201910298019.1A CN201910298019A CN109969181A CN 109969181 A CN109969181 A CN 109969181A CN 201910298019 A CN201910298019 A CN 201910298019A CN 109969181 A CN109969181 A CN 109969181A
- Authority
- CN
- China
- Prior art keywords
- deviation
- layer
- vehicle
- man
- assist torque
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000005457 optimization Methods 0.000 claims abstract description 29
- 238000013461 design Methods 0.000 claims abstract description 21
- 230000007246 mechanism Effects 0.000 claims abstract description 12
- 238000013528 artificial neural network Methods 0.000 claims description 24
- 230000009977 dual effect Effects 0.000 claims description 15
- 238000010606 normalization Methods 0.000 claims description 12
- 210000002569 neuron Anatomy 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 230000001133 acceleration Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 abstract description 4
- 238000012360 testing method Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011112 process operation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
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
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/10—Path keeping
- B60W30/12—Lane keeping
-
- 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
- B60W50/08—Interaction between the driver and the control system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- B60W2510/00—Input parameters relating to a particular sub-units
- B60W2510/20—Steering systems
- B60W2510/202—Steering torque
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/14—Yaw
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2710/00—Output or target parameters relating to a particular sub-units
- B60W2710/20—Steering systems
- B60W2710/202—Steering torque
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Mechanical Engineering (AREA)
- Transportation (AREA)
- Human Computer Interaction (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
The invention discloses a kind of deviation auxiliary system and its deviation householder methods.A personal-machine coordinated control system is arranged in deviation auxiliary system, and man-machine harmony control system includes EPS mechanism, practical assist torque TaOptimization system.EPS mechanism includes that deviation judgment basis obtains module, deviation judgment module, deviates supplementary controlled system starting module.Practical assist torque TaOptimization system include desired orientation disk rotational angle theta*With desired assist torqueModule, man-machine harmony control are obtained according to acquisition module, man-machine harmony controller design module, practical assist torque TaOptimization module.The present invention dynamically adjusts the assist torque of deviation auxiliary system by output auxiliary weight, realize the coordinated control of driver and auxiliary system, it can be while being effectively prevented from vehicle and deviating from lane, reduce interfering with each other between driver and auxiliary system, man-machine conflict is avoided, there is preferable man-machine harmony performance.
Description
The application is that application No. is CN201810031566.9, the applying date 2018/01/12, and entitled one kind
The man-machine harmony control method of deviation auxiliary system and its divisional application of control system.
Technical field
The present invention relates to one of the auxiliary driving technology field of intelligent automobile auxiliary system and its householder methods, especially
It is related to a kind of deviation auxiliary system and its deviation householder method.
Background technique
Deviation auxiliary system (Lane departure assistance system, LDAS) is intelligent automobile auxiliary
The important component of driving technology can assist driver to control vehicle in such a way that active applies and intervenes, thus, such as
What is coordinated the control between driver and auxiliary system and has become the hot spot that domestic and international intelligent automobile auxiliary drives area research
Problem.
There are mainly two types of the approach for realizing deviation auxiliary control: course changing control and differential braking control.Course changing control
Direct torque and corner control can be divided into.Direct torque applies an additional steering force to steering mechanism based on steering system,
To realize auxiliary control;Corner control then needs to control wheel by steering system and goes to desired angle to realize auxiliary control
System.Differential braking control is that desired brake pressure distribution to two sides wheel is carried out differential braking, so that Vehicular yaw responds
Tracking desired value simultaneously realizes deviation auxiliary control.
When carrying out deviation auxiliary using electric power steering, vehicle can realize deviation under various working
Auxiliary has stronger adaptability.However, driver and auxiliary system can be had by carrying out deviation auxiliary using course changing control
Between interfere with each other problem, if coordinate it is inconsistent if will lead to man-machine conflict, this be possible to aggravate pilot control burden,
Influence the safety of automobile transverse direction.Thus, effectively coordinate driver and auxiliary system carries out deviation auxiliary control to be promoted
Man-machine harmony performance is of great significance.
Summary of the invention
Technical problems based on background technology, the invention proposes a kind of deviation auxiliary system and its lane are inclined
From householder method.
Solution of the invention is: a kind of deviation auxiliary system, and a personal-machine coordinated control system, institute is arranged
Stating man-machine harmony control system includes EPS mechanism, practical assist torque TaOptimization system;
EPS mechanism includes that deviation judgment basis obtains module, deviation judgment module, deviates supplementary controlled system
Starting module;The deviation judgment basis obtain module obtain yaw velocity ω, the speed v in vehicle travel process with
And vehicle on road surface relative to the lateral direction of car deviation y of lane center, and it is yaw velocity ω, speed v and vehicle is horizontal
The judgment basis of deviation is carried out as the deviation judgment module to deviation y;The deviation judgment module will
Minimum time needed for prediction wheel touches lane edge as across the road time, the threshold value one of the road Bing Jiangkua time and setting into
Row comparison judges that vehicle will deviate from lane for the moment in the threshold value that is less than the setting;The deviation is auxiliary
Control system starting module is helped to be decided whether to start deviation auxiliary according to the judging result of the deviation judgment module
System;
Practical assist torque TaOptimization system include desired orientation disk rotational angle theta*With desired assist torqueModule is obtained,
Man-machine harmony control is according to acquisition module, man-machine harmony controller design module, practical assist torque TaOptimization module;It is expected that side
To disk rotational angle theta*With desired assist torqueIt obtains module to be used for after deviation auxiliary system starting, according to vehicle cross
To deviation y and destination path f (t), desired orientation disk rotational angle theta needed for obtaining Vehicular turn*, further according to desired orientation disk corner
θ*Obtain desired assist torqueMan-machine harmony control is used to obtain the actual operation torque T of driver according to module is obtainedd,
Torque T will be operateddThe foundation controlled with lateral direction of car deviation y as man-machine harmony;Man-machine harmony controller design module is used for
The man-machine harmony controller for designing the output of dual input list will operate torque TdIt is controlled with lateral direction of car deviation y as man-machine harmony
Two inputs of device, the output of man-machine harmony controller are weight coefficient σ;Practical assist torque TaOptimization module is for passing through power
Weight factor sigma and desired assist torqueProduct is done dynamically to adjust the practical assist torque T of the deviation auxiliary systema's
Size;
Wherein, the principle of the fuzzy neural network controller satisfaction includes:
(1) as | Td| > Td max, vehicle is in a state of emergency at this time, practical assist torque TaWeight coefficient σ it is minimum, drive
The person of sailing fully takes up vehicle driving sovereignty, whereinIt is expressed as judging threshold value two set by driver's mode of operation most
Big value;
(2) as | Td| < Td 0, driver does not operate steering wheel at this time, and the deviation auxiliary system occupies vehicle row
Sovereignty are sailed, weight coefficient σ increases with the increase of lateral direction of car deviation y, whereinThreshold value two set by indicating is most
Small value;
(3) work as Td 0≤|Td|≤Td maxAnd | y | < ymin, vehicle is in lane center at this time, without departing from the danger in lane out
Danger, so to reduce practical assist torque TaWeight coefficient σ, give driver's vehicle driving as much as possible sovereignty, wherein
yminExpression thinks that vehicle is still within threshold value three set by lane center;
(4) work as Td 0≤|Td|≤Td maxAnd | y | >=yminIf operating torque TdWith practical assist torque TaIt is contrary, it says
Bright driver's maloperation is needed at this time to practical assist torque TaWeight coefficient σ is turned up to correct vehicle driving trace;If operation
Torque TdWith practical assist torque TaDirection is identical, and it is correct to illustrate that driver turns to.
As a further improvement of the foregoing solution, desired orientation disk rotational angle theta is calculated by pilot model*, driver
Model uses Single-point preview model: for f (t) for vehicle target track, y (t) is the lateral coordinate of current vehicle position, and T is when taking aim in advance
Between;Desired orientation disk rotational angle theta*Calculation method the following steps are included:
One, assume that preview distance is d, take aim at the relationship between time T and preview distance d in advance are as follows:
According to the side velocity of vehicle, that is, speed v and vehicle side acceleration, the lateral of t+T moment vehicle location is predicted
Coordinate y (t+T) selects a steering angle that vehicle is made to generate side acceleration at this timeIn the side of t+T moment vehicle location
It is equal with lateral coordinate f (t+T) of target trajectory to coordinate y (t+T), then:
F (t+T)=y (t+T)
Two formula of simultaneous can obtain optimal side acceleration
Define practical side accelerationWith the relationship between actual steering wheel rotational angle theta:
In formula, R is motor turning radius, iswIndicate that steering system ratio, L indicate the wheelbase of vehicle;
Two, optimal steering wheel angle needed for obtaining tracking target trajectory it is expected steering wheel angle θ*:
The man-machine harmony controller includes the fuzzy neural network controller based on five etale topology structures, the fuzzy mind
The five etale topology structures through network controller are as follows: input layer, blurring layer, reasoning layer, normalization layer and output layer;Turned with operation
Square TdIt is dual input with lateral direction of car deviation y, weight coefficient σ is single output.
As a further improvement of the foregoing solution, if the operation torque T of inputdDomain be [- 8,8], fuzzy subset is
{ NB, NM, NS, Z, PS, PM, PB }, NB, NM, NS, Z, PS, PM, PB are operation torque TdFuzzy Linguistic Variable after blurring,
Respectively indicate { negative big, to bear, bear small, zero, just small, center is honest };The domain of the lateral direction of car deviation y of input be set as [-
0.6,0.6], fuzzy subset is also { NB, NM, NS, Z, PS, PM, PB }, respectively indicate it is negative big, bear, bear it is small, zero, it is just small, just
In, it is honest };The domain of the weight coefficient σ of output is [0,1], and fuzzy subset is { Z, S, M, L, VL }, respectively indicate zero, it is small,
In, greatly, very greatly };Enable input vector X=[x1,x2]T(x1=Td,x2=y), the output y of kth layer(k), (k=1,2,3,4,5)
It indicates, each layer function are as follows: first layer: input layer, the second layer: blurring layer, third layer: reasoning layer, the 4th layer: normalization layer,
Layer 5: output layer.
Further, first layer: input layer, the corresponding continuous variable x of each neuron node of input layeri, this
Input data is directly transmitted to the second node layer by the node of layer, thus, outputIt is expressed as follows:
The second layer: blurring layer, by the continuous variable x of inputiValue, according to being subordinate on the three of definition fuzzy subsets
Spend function and carry out Fuzzy processing, this layer of each node on behalf a linguistic variable value, total node number 14, first layer i-th
It is a to export corresponding j-th stage degree of membershipCalculation formula indicates are as follows:
In formula: cij,σijRespectively indicate center and the width of membership function;
Third layer: reasoning layer, each neuron node represents a corresponding fuzzy rule, by matching the second node layer
Obtained degree of membership, calculates the relevance grade of every fuzzy rule, total node number n, wherein n=49, then m-th of section of third layer
PointOutput are as follows:
In formula,Corresponding j-th stage degree of membership is exported for first layer the 1st,It is corresponded to for the 2nd output of first layer
J-th stage degree of membership;
4th layer: normalization layer carries out overall normalization to network structure and calculates, total node number n, the 4th layer m-th
NodeOutput are as follows:
Layer 5: the variable sharpening after blurring is carried out Anti-fuzzy calculating by output layer, and network exports y(5)Equal to
4 layers of each node export the product summation of corresponding weight:
In formula: wmIndicate the 4th layer of m-th of node and output nodeBetween connection weight.
As a further improvement of the foregoing solution, by actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*It makes the difference, and leads to
Cross BP neural network PID controller obtain Vehicular turn needed for expectation assist torque
As a further improvement of the foregoing solution, will prediction wheel touch lane edge needed for minimum time as across
The road time compares the threshold value one of across road time and setting, starts for the moment in the threshold value that across the road time is less than the setting
The deviation auxiliary system.
As a further improvement of the foregoing solution, it if calculated across the road time is more than or equal to the threshold value one of setting, says
Bright vehicle will not will deviate from lane, then do not start deviation auxiliary system.
Further, the judgement algorithm using across the road time as deviation is sentenced based on the vehicle deviation across the road time
Disconnected algorithm is touched needed for the edge of lane by the vehicle movement model prediction vehicle driving trace established to calculate wheel
Minimum time.
Preferably, the mode of across road time TLC is calculated are as follows:
In formula, dlaneIndicate lane width, dbIndicate wheelspan, ω be vehicle yaw velocity, θ be vehicle course angle by
Yaw velocity ω integrates to obtain, and L indicates that the wheelbase of vehicle, v are the speed of vehicle.
The present invention also provides a kind of deviation householder methods, are applied in above-mentioned any deviation auxiliary system,
The deviation householder method the following steps are included:
According to lateral direction of car the deviation y and destination path f (t) in vehicle travel process, the phase needed for obtaining Vehicular turn
Hope steering wheel angle θ*;
According to actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*, expectation assist torque needed for obtaining Vehicular turn
Design the man-machine harmony controller of dual input list output, the operation torque T in vehicle travel processdAnd lateral direction of car
Two inputs of the deviation y as man-machine harmony controller, the output of man-machine harmony controller is weight coefficient σ;
Pass through weight coefficient σ and desired assist torqueDo the reality that product carrys out deviation auxiliary system described in dynamic optimization
Border assist torque TaSize.
Deviation auxiliary system of the invention, it is theoretical based on Fuzzy Neural-network Control, it was assisted for deviation
Man-machine harmony problem in journey between driver and auxiliary system devises the people for considering driver's torque and lateral direction of car deviation
Machine tuning controller.Man-machine harmony controller is turned by the auxiliary that output auxiliary weight dynamically adjusts deviation auxiliary system
Square realizes the coordinated control of driver and auxiliary system.The present invention can while being effectively prevented from vehicle and deviating from lane,
Reduce interfering with each other between driver and auxiliary system, avoids man-machine conflict, there is preferable man-machine harmony performance.
Detailed description of the invention
Fig. 1 is the flow chart of the man-machine harmony control method of deviation auxiliary system of the invention.
Fig. 2 is the structural schematic diagram using the man-machine harmony control system of man-machine harmony control method in Fig. 1.
Fig. 3 is the Single-point preview model schematic that pilot model uses in Fig. 2.
Fig. 4 is the control structure figure of PID controller in Fig. 2.
Fig. 5 is the fuzzy neural network topological structure schematic diagram of tuning controller in Fig. 2.
Fig. 6 is the practical assist torque T of deviation auxiliary system of the inventionaOptimization method flow chart.
Fig. 7 is the hardware-in-the-loop test flow diagram of man-machine harmony control system in Fig. 2.
Fig. 8 is the operation torque T of the i.e. driver of driver's input torque of man-machine harmony control system in Fig. 2dTest
Result curve figure.
Fig. 9 is the test result curve graph of the weight coefficient σ of man-machine harmony control system in Fig. 2.
Figure 10 is the practical assist torque T of man-machine harmony control system in Fig. 2aTest result curve graph.
Figure 11 is the test result curve graph of the lateral direction of car deviation y of man-machine harmony control system in Fig. 2.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Traditional deviation auxiliary system judges that vehicle will deviate from lane and driver does not operate steering wheel working as
When, it will enable, once driver intervenes, auxiliary system will stop working.System passes through electric booster turning mechanism, that is, EPS
(Electric Power steering system) carries out deviation auxiliary.The motor of EPS is such as driven to apply to steering column
Torque changes vehicle front corner δf, vehicle front corner δfChange cause the adjustment of vehicle-state and position, be embodied in vehicle
Adjustment of the vehicle on road surface relative to the lateral direction of car deviation y of lane center in driving process.
The man-machine harmony control method of deviation auxiliary system of the invention is used for when vehicle will deviate from lane,
Collaboration driver completes to turn to jointly.The system can effectively coordinate driver and deviation auxiliary system, carry out in due course
Deviation auxiliary control is to promote man-machine harmony performance.Thus, the present invention can deviate from lane being effectively prevented from vehicle
While, reduce interfering with each other between driver and deviation auxiliary system, avoid man-machine conflict, there is preferable man-machine association
Tonality energy.
Embodiment 1
Fig. 1 and Fig. 2 is please referred to, the man-machine harmony control method of deviation auxiliary system of the invention includes following step
Suddenly.
Step S11, obtain vehicle travel process in yaw velocity ω, speed v and vehicle on road surface relative to
The lateral direction of car deviation y of lane center, and using yaw velocity ω, speed v and lateral direction of car deviation y as deviation
Judgment basis.
Step S12, minimum time needed for prediction wheel is touched lane edge, will across the road times as across the road time
It is compared with the threshold value one of setting, judges that vehicle will deviate from for the moment in the threshold value that is less than the setting
Lane.
In the present embodiment, the judgement algorithm using across the road time as deviation.By calculated across the road time and
The threshold value one of setting compares, and then judges whether vehicle will deviate from lane.
Deviateed based on the vehicle across the road time and judge that algorithm passes through the vehicle movement model prediction vehicle driving trace established,
To calculate the minimum time Ji Kua road time needed for wheel touches lane edge.Calculate the specific table of across road time TLC
Up to formula are as follows:
In formula, dlaneIndicate lane width, dbIndicate that wheelspan, θ are vehicle course angle (i.e. actual steering wheel corner), it can be by
Yaw velocity ω integrates to obtain, and L indicates that wheelbase, the yaw velocity ω, speed v, vehicle that ω, v, y are all from step S11 are horizontal
To deviation y.
Step S13 decides whether to start deviation auxiliary system according to judging result.
When judging that vehicle will deviate from lane, start the deviation auxiliary system.If in step S12, meter
Calculated is less than the threshold value one of setting, illustrates that vehicle will deviate from lane, then it is auxiliary to start deviation by step S13
Auxiliary system.If calculated across the road time is more than or equal to the threshold value one of setting, illustrate that vehicle will not will deviate from lane, then
Deviation auxiliary system is not started.
Step S14, according to lateral direction of car deviation y and actual steering wheel rotational angle theta, desired orientation needed for obtaining Vehicular turn
Disk rotational angle theta*With desired assist torque
In the present embodiment, according to the state parameters such as lateral direction of car deviation y and actual steering wheel rotational angle theta, pass through driver
Desired orientation disk rotational angle theta needed for model and the pid algorithm of neural network obtain Vehicular turn respectively*With desired assist torqueIt first passes through pilot model and calculates desired orientation disk rotational angle theta*, by actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*
It makes the difference, and expectation assist torque needed for obtaining Vehicular turn by the PID controller of BP neural network
Pilot model is Single-point preview model as shown in Figure 3: f (t) is vehicle target track, and y (t) is that vehicle is current
The lateral coordinate in position, T are to take aim at the time in advance.
Assuming that preview distance is d, the relationship between time T and preview distance d is taken aim in advance are as follows:
According to the side velocity of vehicle, that is, speed v and vehicle side acceleration, t+T moment vehicle location can be predicted
Lateral coordinate y (t+T), selects an ideal steering angle that vehicle is made to generate side acceleration at this timeAt the t+T moment
The lateral coordinate y (t+T) of vehicle location is equal with lateral coordinate f (t+T) of target trajectory, then can obtain:
F (t+T)=y (t+T)
Two formula of simultaneous can obtain optimal side acceleration
According to vehicle kinematics relationship, available practical side accelerationWith the pass between actual steering wheel rotational angle theta
System:
In formula, R is motor turning radius, iswIndicate steering system ratio.
Optimal steering wheel angle needed for finally obtaining tracking target trajectory it is expected steering wheel angle θ*:
The PID controller of BP neural network is as shown in figure 4, i.e. Neural Network PID Control structure is mainly controlled by classical PID
Device processed and neural network two parts are constituted.Classical PID controller: directly to controlled device carry out closed-loop control, the three of controller
A parameter is on-line tuning.Neural network: the output state of its output layer neuron corresponds to three adjustable ginsengs of PID controller
Number, self study and adjustment weighting coefficient by neural network, so that the output of neural network corresponds to certain optimal control law
Under pid control parameter.
Neural network uses three layers of feedforward network of 3-5-3 structure.The number of input layer is 3, respectively sideway
Angular speed desired value, actual value and deviation;Hidden layer neuron number is 5;Output layer neuron number is 3, i.e. PID control is joined
Number.
Enable input vector X=[x1(n),x2(n),x3(n)]T, x1(n),x2(n),x3(n) ω is respectively indicated*(n),ω(n)
And its deviation e (n);The output y of kth layer(k)(n), (k=1,2,3) is indicated;The activation primitive of hidden layer neuron takes positive and negative
Symmetrical Sigmoid function:
Output layer exports
Since these three parameters cannot be negative, so the activation primitive of output layer is
Therefore, the control law of BP neural network PID controller is
Defining performance index function is
As shown in figure 5, being iterated amendment to network weights coefficient using BP learning algorithm, i.e., by ε (n) to weighting coefficient
Negative gradient direction search for adjustment, and additional one momentum term for making to search for fast convergence global minimal
In formula, η is learning rate;α is factor of momentum;wliFor the weighting coefficient of hidden layer and output layer.
Step S15 obtains the actual operation torque T of driverd, torque T will be operateddWith lateral direction of car deviation y as man-machine
The foundation of coordinated control.
Step S16, the man-machine harmony controller of design dual input list output, operates torque TdMake with lateral direction of car deviation y
For two inputs of man-machine tuning controller, the output of man-machine harmony controller is weight coefficient σ.That is, according to operation torque Td
With the man-machine harmony controller of lateral direction of car deviation y design dual input list output.
The man-machine harmony controller includes the fuzzy neural network controller based on five etale topology structures, the fuzzy mind
The five etale topology structures through network controller are as follows: input layer, blurring layer, reasoning layer, normalization layer and output layer;Turned with operation
Square TdIt is dual input with lateral direction of car deviation y, weight coefficient σ is single output.Therefore the fuzznet based on five etale topology structures
The man-machine harmony controller of network Theoretical Design dual input list output.
The man-machine harmony controller is based on Fuzzy Neural Network Theory and fully considers that driver operates torque TdAnd vehicle
Lateral deviation y and design.
Fuzzy neural network controller for man-machine harmony design needs meet principle specifically includes.
(1) when driver's torque | Td| > Td max, vehicle is in a state of emergency at this time, practical assist torque TaWeight system
Number is minimum, and driver fully takes up the sovereignty of vehicle driving.
(2) as | Td| < Td 0, driver does not operate steering wheel at this time, and the deviation auxiliary system occupies vehicle row
Sovereignty are sailed, weight coefficient σ increases with the increase of lateral lateral direction of car deviation y.Wherein,Judgement is expressed as to drive
The maximum value and minimum value of threshold value two set by member's mode of operation.
(3) work as Td 0≤|Td|≤Td maxAnd | y | < ymin, vehicle is in lane center at this time, without departing from the danger in lane out
Danger, so to reduce practical assist torque TaWeight coefficient σ, give driver's vehicle driving as much as possible sovereignty.Wherein,
yminExpression thinks that vehicle is still within threshold value three set by lane center.
(4) work as Td 0≤|Td|≤Td maxAnd | y | >=ymin, the three kinds of situation discussion of this time-division: turn if driver's torque operates
Square TdWith practical assist torque TaIt is contrary, illustrate driver's maloperation, is needed at this time to practical assist torque TaBiggish power
Weight factor sigma is to correct vehicle driving trace;If operating torque TdWith practical assist torque TaDirection is identical, illustrates that driver turns to
Correctly.Driver's torque is bigger, practical assist torque TaWeight coefficient σ with regard to smaller, to reduce auxiliary system to driver's
Intervene;If lateral deviation y is larger, practical assist torque TaWeight coefficient σ it is also larger, vice versa.
The fuzzy neural network of designed man-machine harmony controller uses dual input/mono- five etale topology structures exported,
That is input layer, blurring layer, reasoning layer, normalization layer and output layer.To operate torque TdIt is input with lateral direction of car deviation y,
Weight coefficient σ is output.
If the operation torque T of inputdDomain be [- 8,8], fuzzy subset be { NB, NM, NS, Z, PS, PM, PB }, respectively
Indicate { negative big, to bear, bear small, zero, just small, center is honest };The domain of vehicle lateral deviation y is set as [- 0.6,0.6], obscures
Subset is also { NB, NM, NS, Z, PS, PM, PB }, respectively indicates { negative big, to bear, bear small, zero, just small, center is honest };Output
Weight coefficient σ domain be [0,1], fuzzy subset be { Z, S, M, L, VL }, respectively indicate zero, it is small, in, greatly, very greatly.It enables
Input vector X=[x1,x2]T(x1=Td,x2=y), the output y of kth layer(k), (k=1,2,3,4,5) is indicated, each layer function
It is as follows:
First layer: input layer.The corresponding continuous variable x of each neuron node of input layeri, the node of this layer is straight
It connects and input data is transmitted to the second node layer, thus, outputIt is expressed as follows:
The second layer: blurring layer.By the continuous variable x of inputiValue according to the degree of membership letter on the fuzzy subset of definition
Number carries out Fuzzy processings, this layer of each node on behalf a linguistic variable value, total node number 14.1st layer of i-th of output
Corresponding j-th stage degree of membershipCalculation formula may be expressed as:
In formula: cij,σijRespectively indicate center and the width of membership function.
Third layer: reasoning layer.Each neuron node represents a corresponding fuzzy rule, is obtained by matching the 2nd layer
Degree of membership, calculate the relevance grade of every rule.Total node number is n (n=49), then m-th of nodeOutput are as follows:
In formula,Corresponding j-th stage degree of membership is exported for first layer the 1st,It is corresponded to for the 2nd output of first layer
J-th stage degree of membership.It is briefly exactly the output of the second layer when i is respectively 1 and 2.
4th layer: normalization layer.Carry out overall normalization to network structure to calculate, total node number n, the 4th layer m-th
NodeOutput are as follows:
Layer 5: output layer.By the variable sharpening after blurring, Anti-fuzzy calculating is carried out.Network exports y(5)Equal to
4 layers of each node export the product summation of corresponding weight.
In formula: wmIndicate the 4th layer of m-th of node and output nodeBetween connection weight.
Step S17 passes through weight coefficient σ and desired assist torqueProduct is done dynamically to adjust the deviation auxiliary
The practical assist torque T of systemaSize.
Man-machine harmony controller is according to operation torque TdA weight coefficient σ is generated in real time with the value of lateral direction of car deviation y,
And practical assist torque T is dynamically adjusted by this weight coefficient σaSize, coordinate driver while ensuring safety
Control between auxiliary system;
Designed man-machine harmony controller is according to the operation torque T of driverdIt is produced in real time with the value of lateral direction of car deviation y
A raw dynamic weight coefficient σ, and pass through expectation assist torque needed for this weight coefficient σ and Vehicular turnDo product
To adjust practical assist torque T in real timeaSize, not only can guarantee vehicle without departing from lane out but also realize driver and auxiliary system
Coordinated control between system.
The practical assist torque T obtained through the above stepsaWith the operation torque T of driverdCollective effect is in steering system
System, if driver's torque operates torque TdWith practical assist torque TaIt is contrary, illustrate driver's maloperation, needs at this time
To practical assist torque TaBiggish weight coefficient σ is to correct vehicle driving trace.Lane can be individually carried out by EPS system
Deviate auxiliary, such as changes vehicle front corner δf, vehicle front corner δfChange cause the adjustment of vehicle-state, it is final to change
Lateral direction of car deviation y.
If operating torque TdWith practical assist torque TaDirection is identical, and it is correct to illustrate that driver turns to.Without passing through EPS machine
Structure carries out deviation auxiliary.Operate torque TdIt is bigger, practical assist torque TaWeight coefficient σ with regard to smaller, with reduce auxiliary system
The intervention united to driver, at this point, the assist torque Collaborative Control Vehicular turn that the operation of driver and auxiliary system provide.If
Lateral direction of car deviation y is larger, practical assist torque TaWeight coefficient σ it is also larger, vice versa.
In other embodiments, the man-machine harmony control method of deviation auxiliary system of the invention, it may include following
Simplify step:
Minimum time needed for prediction wheel is touched lane edge is as across the road time, by across road time and setting
Threshold value one compares, and starts the deviation auxiliary system for the moment in the threshold value that across the road time is less than the setting;
According to lateral direction of car deviation y and destination path f (t), desired orientation disk rotational angle theta needed for obtaining Vehicular turn*;
According to desired orientation disk rotational angle theta*Obtain desired assist torque
Design the actual operation torque T of driverdWith lateral direction of car deviation y as dual input, weight coefficient σ as single defeated
Man-machine harmony controller out;
Pass through weight coefficient σ and desired assist torqueProduct is done dynamically to adjust the reality of the deviation auxiliary system
Border assist torque TaSize.
The method that present embodiment is proposed is intended to provide a kind of man-machine harmony control method of deviation auxiliary system,
This method is for the driver in deviation supporting process and the man-machine harmony problem between deviation auxiliary system, application
Fuzzy Neural-network Control is theoretical, and design considers the operation torque T of driverdMan-machine harmony with lateral direction of car deviation y controls
Device dynamically adjusts the practical assist torque T of deviation auxiliary system by output auxiliary weight coefficient σa, realize driver
With the coordinated control of auxiliary system.The present invention can while being effectively prevented from vehicle and deviating from lane, reduce driver and
Interfering with each other between auxiliary system avoids man-machine conflict, there is preferable man-machine harmony performance, can further genralrlization.
Embodiment 2
Show the man-machine harmony control using the man-machine harmony control method of embodiment 1 referring to Fig. 2, Fig. 2
The structural schematic diagram of system.Man-machine harmony control system of the invention includes EPS mechanism, practical assist torque TaOptimization system
System.
EPS mechanism includes that deviation judgment basis obtains module, deviation judgment module, deviates supplementary controlled system
Starting module.
The deviation judgment basis obtain module obtain yaw velocity ω, the speed v in vehicle travel process with
And vehicle on road surface relative to the lateral direction of car deviation y of lane center, and it is yaw velocity ω, speed v and vehicle is horizontal
The judgment basis of deviation is carried out as the deviation judgment module to deviation y.
The deviation judgment module will predict wheel touch lane edge needed for minimum time as it is across road when
Between, the threshold value one of the road Bing Jiangkua time and setting compares, and sentences for the moment in the threshold value that is less than the setting
Disconnected vehicle will deviate from lane.
The deviation supplementary controlled system starting module is determined according to the judging result of the deviation judgment module
No starting deviation auxiliary system.
Practical assist torque TaOptimization system include desired orientation disk rotational angle theta*With desired assist torqueModule is obtained,
Man-machine harmony control is according to acquisition module, man-machine harmony controller design module, practical assist torque TaOptimization module.
Desired orientation disk rotational angle theta*With desired assist torqueModule is obtained, according to lateral direction of car deviation y and destination path f
(t), desired orientation disk rotational angle theta needed for obtaining Vehicular turn*With desired assist torque
Man-machine harmony control obtains the actual operation torque T of driver according to module is obtainedd, torque T will be operateddAnd vehicle
The foundation that lateral deviation y is controlled as man-machine harmony.
The man-machine harmony controller of man-machine harmony controller design module design dual input list output, operates torque TdAnd vehicle
Two inputs of lateral deviation y as man-machine harmony controller, the output of man-machine harmony controller is weight coefficient σ.
Practical assist torque TaOptimization module passes through weight coefficient σ and desired assist torqueProduct is done dynamically to adjust institute
State the practical assist torque T of deviation auxiliary systemaSize.
The details of man-machine harmony control system describes in the man-machine harmony control method of embodiment 1, no longer tired herein
It states.
Embodiment 3
Fig. 2, Fig. 6 are please referred to, the present embodiment 3 illustrates the practical assist torque T of deviation auxiliary system of the inventiona
Optimization method, the optimization method includes the following steps.
Step S21 obtains Vehicular turn according to lateral direction of car the deviation y and destination path f (t) in vehicle travel process
Required desired orientation disk rotational angle theta*。
According to lateral direction of car deviation y and destination path f (t), desired orientation disk rotational angle theta is calculated by pilot model*,
Desired orientation disk rotational angle theta*Calculation method as described by the step S14 in embodiment 1, be not repeated introduction herein.
Step S22, according to actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*, expectation needed for obtaining Vehicular turn
Assist torque
By actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*It makes the difference, and is obtained by the PID controller of BP neural network
Expectation assist torque needed for Vehicular turn outIt is expected that assist torqueCalculation method such as embodiment 1 in step S14
It is described, it is not repeated introduction herein.
Step S23 designs the man-machine harmony controller of dual input list output, the operation torque T in vehicle travel processdWith
Two inputs of the lateral direction of car deviation y as man-machine harmony controller, the output of man-machine harmony controller is weight coefficient σ.
The calculation method of weight coefficient σ is not repeated introduction as described by the step S16 in embodiment 1 herein.
Step S24 passes through weight coefficient σ and desired assist torqueIt does product and carrys out the auxiliary of deviation described in dynamic optimization
The practical assist torque T of systemaSize.
If driver's torque operates torque TdWith practical assist torque TaIt is contrary, illustrate driver's maloperation, at this time
It needs to practical assist torque TaBiggish weight coefficient σ is to correct vehicle driving trace.It can individually be carried out by EPS system
Deviation auxiliary, such as changes vehicle front corner δf, vehicle front corner δfChange cause the adjustment of bus or train route model, finally
Change lateral direction of car deviation y.
If operating torque TdWith practical assist torque TaDirection is identical, and it is correct to illustrate that driver turns to.Without passing through EPS machine
Structure carries out deviation auxiliary.Operate torque TdIt is bigger, practical assist torque TaWeight coefficient σ with regard to smaller, with reduce auxiliary system
The intervention united to driver, at this point, the operation of driver synchronous can be carried out with the deviation of EPS mechanism auxiliary.If vehicle
Lateral deviation y is larger, practical assist torque TaWeight coefficient σ it is also larger, vice versa.
Embodiment 4
The practical assist torque T using embodiment 3 is also illustrated referring to Fig. 2, Fig. 2aOptimization method reality
Assist torque TaOptimization system structural schematic diagram.Practical assist torque T of the inventionaOptimization system include desired orientation
Disk rotational angle theta*Module is obtained, it is expected that assist torqueObtain module, man-machine harmony controller design module, practical assist torque Ta
Optimization module.
Desired orientation disk rotational angle theta*Module is obtained according to the lateral direction of car deviation y and destination path f in vehicle travel process
(t), desired orientation disk rotational angle theta needed for obtaining Vehicular turn*。
It is expected that assist torqueModule is obtained according to actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*, obtain vehicle
Expectation assist torque needed for turning to
The man-machine harmony controller of man-machine harmony controller design module design dual input list output, in vehicle travel process
Operation torque TdTwo inputs with lateral direction of car deviation y as man-machine harmony controller, the output of man-machine harmony controller
For weight coefficient σ.
Practical assist torque TaOptimization module passes through weight coefficient σ and desired assist torqueIt does product and comes dynamic optimization institute
State the practical assist torque T of deviation auxiliary systemaSize.
Practical assist torque TaOptimization system details in the practical assist torque T of embodiment 3aOptimization method in
Description, is not repeated herein.
Embodiment 5
For verifying embodiment 1 in man-machine harmony control method validity and feasibility, below in conjunction with specifically to man-machine association
Control method is verified.
Using the simulated environment based on CarSim auto model, combines LabVIEW and carry out hardware-in-the-loop test research.Test
Platform and test block diagram are as shown in Figure 7.The testing stand that the present invention is built mainly by host computer, slave computer, interface system and turns
It is formed to several parts of system.CarSim Full Vehicle Dynamics model and virtual road are established according to vehicle parameter in host computer, joined
CarSim/LabVIEW is closed, LabVIEW deviation auxiliary control program is write;Slave computer is the PXI system of NI, real time execution
The program that host computer is established;Interface system is the signals such as the collected torque of sensor to be transmitted to PXI system, while will control
Signal is exported to the controller of executing agency (such as the EPS motor controller of control assist torque and watching for generation steering response
Take motor).
Select forthright to emulate road, have a lot of social connections 3.75m, and constant speed is 80km/h, applies turning for 10Nm in 1s-1.5s
Square makes automotive run-off-road center, chooses two kinds of representative driver's modes of operation and carries out man-machine harmony control strategy
Verification experimental verification, i.e., in automotive run-off-road, driver reacts, and carries out maloperation and correct operation.
Fig. 8-Figure 11 is man-machine coordination control strategy test result, and wherein Fig. 8 is driver's input torque, that is, driver
Operate torque TdTest result curve graph, Fig. 9 be weight coefficient σ test result curve graph, Figure 10 be practical assist torque Ta
Test result curve graph, Figure 11 be lateral direction of car deviation y test result curve graph.
When driver's steering is correct, the output weight coefficient σ of man-machine harmony controller is decreased obviously, practical assist torque
TaAlso relatively small, thus given driver more with sovereign right, reduce interference of the auxiliary system to driver.When driver misses
When operating steering wheel, output weight is maintained at the larger value, and pilot controller, that is, EPS mechanism exports biggish practical assist torque Ta
To make up the operation torque T that driver applies mistaked.From fig. 10 it can be seen that no matter what driver carries out when vehicle deviates
Kind operation, LDAS, that is, deviation auxiliary system can still guarantee that vehicle does not deflect away from lane.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of deviation auxiliary system, which is characterized in that a personal-machine coordinated control system, the man-machine harmony is arranged in it
Control system includes EPS mechanism, practical assist torque TaOptimization system;
EPS mechanism includes that deviation judgment basis obtains module, deviation judgment module, deviates supplementary controlled system starting
Module;The deviation judgment basis obtains yaw velocity ω, speed v and the vehicle in module acquisition vehicle travel process
Relative to the lateral direction of car deviation y of lane center on road surface, and yaw velocity ω, speed v and lateral direction of car is inclined
Poor y carries out the judgment basis of deviation as the deviation judgment module;The deviation judgment module will be predicted
Minimum time needed for wheel touches lane edge is used as across the road time, and the threshold value one of the road Bing Jiangkua time and setting carries out pair
Than judging that vehicle will deviate from lane for the moment in the threshold value that is less than the setting;The deviation auxiliary control
System starting module processed decides whether to start deviation auxiliary system according to the judging result of the deviation judgment module;
Practical assist torque TaOptimization system include desired orientation disk rotational angle theta*With desired assist torqueObtain module, man-machine
Coordinated control is according to acquisition module, man-machine harmony controller design module, practical assist torque TaOptimization module;Desired orientation disk
Rotational angle theta*With desired assist torqueModule is obtained to be used for after deviation auxiliary system starting, it is inclined according to lateral direction of car
Poor y and destination path f (t), desired orientation disk rotational angle theta needed for obtaining Vehicular turn*, further according to desired orientation disk rotational angle theta*?
Assist torque it is expected outMan-machine harmony control is used to obtain the actual operation torque T of driver according to module is obtainedd, will grasp
Make torque TdThe foundation controlled with lateral direction of car deviation y as man-machine harmony;Man-machine harmony controller design module is for designing
The man-machine harmony controller of dual input list output will operate torque TdWith lateral direction of car deviation y as man-machine harmony controller
Two inputs, the output of man-machine harmony controller are weight coefficient σ;Practical assist torque TaOptimization module is used to pass through weight system
Number σ and desired assist torqueProduct is done dynamically to adjust the practical assist torque T of the deviation auxiliary systemaIt is big
It is small;
Wherein, the principle of the fuzzy neural network controller satisfaction includes:
(1) as | Td| > Td max, vehicle is in a state of emergency at this time, practical assist torque TaWeight coefficient σ it is minimum, driver
Fully take up vehicle driving sovereignty, whereinIt is expressed as judging the maximum value of threshold value two set by driver's mode of operation;
(2) as | Td| < Td 0, driver does not operate steering wheel at this time, and the deviation auxiliary system occupies vehicle driving master
Power, weight coefficient σ increase with the increase of lateral direction of car deviation y, whereinIndicate the minimum value of set threshold value two;
(3) work as Td 0≤|Td|≤Td maxAnd | y | < ymin, vehicle is in lane center at this time, without departing from the danger in lane out, institute
To reduce practical assist torque TaWeight coefficient σ, give driver's vehicle driving as much as possible sovereignty, wherein yminIt indicates
Think that vehicle is still within threshold value three set by lane center;
(4) work as Td 0≤|Td|≤Td maxAnd | y | >=yminIf operating torque TdWith practical assist torque TaContrary, explanation is driven
The person's of sailing maloperation is needed at this time to practical assist torque TaWeight coefficient σ is turned up to correct vehicle driving trace;If operating torque
TdWith practical assist torque TaDirection is identical, and it is correct to illustrate that driver turns to.
2. deviation auxiliary system as described in claim 1, which is characterized in that calculate expectation side by pilot model
To disk rotational angle theta*, pilot model uses Single-point preview model: for f (t) for vehicle target track, y (t) is current vehicle position side
To coordinate, T is to take aim at the time in advance;Desired orientation disk rotational angle theta*Calculation method the following steps are included:
One, assume that preview distance is d, take aim at the relationship between time T and preview distance d in advance are as follows:
According to the side velocity of vehicle, that is, speed v and vehicle side acceleration, the lateral coordinate of t+T moment vehicle location is predicted
Y (t+T) selects a steering angle that vehicle is made to generate side acceleration at this timeIn the lateral seat of t+T moment vehicle location
Mark y (t+T) is equal with lateral coordinate f (t+T) of target trajectory, then:
F (t+T)=y (t+T)
Two formula of simultaneous can obtain optimal side acceleration
Define practical side accelerationWith the relationship between actual steering wheel rotational angle theta:
In formula, R is motor turning radius, iswIndicate that steering system ratio, L indicate the wheelbase of vehicle;
Two, optimal steering wheel angle needed for obtaining tracking target trajectory it is expected steering wheel angle θ*:
The man-machine harmony controller includes the fuzzy neural network controller based on five etale topology structures, the fuzznet
Five etale topology structures of network controller are as follows: input layer, blurring layer, reasoning layer, normalization layer and output layer;To operate torque Td
It is dual input with lateral direction of car deviation y, weight coefficient σ is single output.
3. deviation auxiliary system as described in claim 1, which is characterized in that set the operation torque T of inputdDomain be
[- 8,8], fuzzy subset are { NB, NM, NS, Z, PS, PM, PB }, and NB, NM, NS, Z, PS, PM, PB is operation torque TdBlurring
Fuzzy Linguistic Variable afterwards respectively indicates { negative big, to bear, bear small, zero, just small, center is honest };The lateral direction of car of input is inclined
The domain of poor y is set as [- 0.6,0.6], and fuzzy subset is also { NB, NM, NS, Z, PS, PM, PB }, respectively indicate negative big, it bears,
Bear it is small, zero, it is just small, center, it is honest;The domain of the weight coefficient σ of output is [0,1], and fuzzy subset is { Z, S, M, L, VL },
Respectively indicate zero, it is small, in, greatly, very greatly };Enable input vector X=[x1,x2]T(x1=Td,x2=y), the output y of kth layer(k), (k=1,2,3,4,5) is indicated, each layer function are as follows: first layer: input layer, the second layer: blurring layer, third layer: reasoning layer,
4th layer: normalization layer, layer 5: output layer.
4. deviation auxiliary system as claimed in claim 3, which is characterized in that first layer: input layer, each of input layer
Neuron node corresponds to a continuous variable xi, input data is directly transmitted to the second node layer by the node of this layer, thus, defeated
OutIt is expressed as follows:
The second layer: blurring layer, by the continuous variable x of inputiValue, according to the degree of membership letter on the three of definition fuzzy subsets
Number carries out Fuzzy processings, this layer of each node on behalf a linguistic variable value, total node number 14, i-th of first layer is defeated
Corresponding j-th stage degree of membership outCalculation formula indicates are as follows:
In formula: cij,σijRespectively indicate center and the width of membership function;
Third layer: reasoning layer, each neuron node represent a corresponding fuzzy rule, are obtained by matching the second node layer
Degree of membership, calculate the relevance grade of every fuzzy rule, total node number n, wherein n=49, then m-th of node of third layer
Output are as follows:
In formula,Corresponding j-th stage degree of membership is exported for first layer the 1st,For the corresponding jth of first layer the 2nd output
Grade degree of membership;
4th layer: normalization layer carries out overall normalization to network structure and calculates, total node number n, the 4th layer of m-th of nodeOutput are as follows:
Layer 5: the variable sharpening after blurring is carried out Anti-fuzzy calculating by output layer, and network exports y(5)Equal to the 4th layer
Each node exports the product summation of corresponding weight:
In formula: wmIndicate the 4th layer of m-th of node and output nodeBetween connection weight.
5. deviation auxiliary system as described in claim 1, which is characterized in that by actual steering wheel rotational angle theta and expectation side
To disk rotational angle theta*It makes the difference, and expectation assist torque needed for obtaining Vehicular turn by the PID controller of BP neural network
6. deviation auxiliary system as described in claim 1, which is characterized in that prediction wheel is touched lane edge institute
The minimum time needed is used as across the road time, and the threshold value one of across road time and setting is compared, and is less than in across the road time described
The threshold value of setting starts the deviation auxiliary system for the moment.
7. deviation auxiliary system as described in claim 1, which is characterized in that if calculated across the road time be greater than etc.
In the threshold value one of setting, illustrates that vehicle will not will deviate from lane, then do not start deviation auxiliary system.
8. deviation auxiliary system as claimed in claim 6, which is characterized in that using across the road time as deviation
Judge algorithm, is deviateed based on the vehicle across the road time and judge that algorithm passes through the vehicle movement model prediction vehicle driving rail of foundation
Mark, to calculate minimum time needed for wheel touches lane edge.
9. deviation auxiliary system as claimed in claim 8, which is characterized in that calculate the mode of across road time TLC are as follows:
In formula, dlaneIndicate lane width, dbIndicate that wheelspan, ω are the yaw velocity of vehicle, θ is vehicle course angle by sideway
Angular velocity omega integrates to obtain, and L indicates that the wheelbase of vehicle, v are the speed of vehicle.
10. it is auxiliary to be applied to deviation as in one of claimed in any of claims 1 to 9 for a kind of deviation householder method
In auxiliary system, which is characterized in that the deviation householder method the following steps are included:
According to lateral direction of car the deviation y and destination path f (t) in vehicle travel process, expectation side needed for obtaining Vehicular turn
To disk rotational angle theta*;
According to actual steering wheel rotational angle theta and desired orientation disk rotational angle theta*, expectation assist torque needed for obtaining Vehicular turn
Design the man-machine harmony controller of dual input list output, the operation torque T in vehicle travel processdWith lateral direction of car deviation y
As two inputs of man-machine harmony controller, the output of man-machine harmony controller is weight coefficient σ;
Pass through weight coefficient σ and desired assist torqueDo product come deviation auxiliary system described in dynamic optimization reality it is auxiliary
Help torque TaSize.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910298019.1A CN109969181B (en) | 2018-01-12 | 2018-01-12 | Lane departure auxiliary system and lane departure auxiliary method thereof |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910298019.1A CN109969181B (en) | 2018-01-12 | 2018-01-12 | Lane departure auxiliary system and lane departure auxiliary method thereof |
CN201810031566.9A CN107972667B (en) | 2018-01-12 | 2018-01-12 | A kind of man-machine harmony control method of deviation auxiliary system |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810031566.9A Division CN107972667B (en) | 2018-01-12 | 2018-01-12 | A kind of man-machine harmony control method of deviation auxiliary system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109969181A true CN109969181A (en) | 2019-07-05 |
CN109969181B CN109969181B (en) | 2020-06-05 |
Family
ID=62005873
Family Applications (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810031566.9A Active CN107972667B (en) | 2018-01-12 | 2018-01-12 | A kind of man-machine harmony control method of deviation auxiliary system |
CN201910298019.1A Active CN109969181B (en) | 2018-01-12 | 2018-01-12 | Lane departure auxiliary system and lane departure auxiliary method thereof |
CN201910298010.0A Active CN109969180B (en) | 2018-01-12 | 2018-01-12 | Man-machine coordination control system of lane departure auxiliary system |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810031566.9A Active CN107972667B (en) | 2018-01-12 | 2018-01-12 | A kind of man-machine harmony control method of deviation auxiliary system |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910298010.0A Active CN109969180B (en) | 2018-01-12 | 2018-01-12 | Man-machine coordination control system of lane departure auxiliary system |
Country Status (1)
Country | Link |
---|---|
CN (3) | CN107972667B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111158377A (en) * | 2020-01-15 | 2020-05-15 | 浙江吉利汽车研究院有限公司 | Transverse control method and system for vehicle and vehicle |
CN111175056A (en) * | 2020-01-17 | 2020-05-19 | 金龙联合汽车工业(苏州)有限公司 | Hardware-in-loop test device of commercial vehicle lane keeping system |
CN113120079A (en) * | 2020-01-15 | 2021-07-16 | 通用汽车环球科技运作有限责任公司 | Steering wheel angle offset correction for autonomous vehicles using angle control |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108725453A (en) * | 2018-06-11 | 2018-11-02 | 南京航空航天大学 | Control system and its switch mode are driven altogether based on pilot model and manipulation the man-machine of inverse dynamics |
CN109177974B (en) * | 2018-08-28 | 2020-01-03 | 清华大学 | Man-machine co-driving type lane keeping auxiliary method for intelligent automobile |
CN109760677B (en) * | 2019-03-13 | 2020-09-11 | 广州小鹏汽车科技有限公司 | Lane keeping auxiliary method and system |
CN111923919B (en) * | 2019-05-13 | 2021-11-23 | 广州汽车集团股份有限公司 | Vehicle control method, vehicle control device, computer equipment and storage medium |
CN110329255B (en) * | 2019-07-19 | 2020-11-13 | 中汽研(天津)汽车工程研究院有限公司 | Lane departure auxiliary control method based on man-machine cooperation strategy |
CN111152776B (en) * | 2020-01-10 | 2021-03-23 | 合肥工业大学 | Steering and braking coordination control method and system for unmanned formula racing car |
CN112677991B (en) * | 2020-12-11 | 2022-06-07 | 武汉格罗夫氢能汽车有限公司 | Hydrogen energy automobile lane departure prevention device |
GB2604321A (en) * | 2020-12-31 | 2022-09-07 | Zf Automotive Uk Ltd | Steer |
GB2602477A (en) * | 2020-12-31 | 2022-07-06 | Zf Automotive Uk Ltd | Automotive vehicle control circuit |
GB2602478A (en) * | 2020-12-31 | 2022-07-06 | Zf Automotive Uk Ltd | Motor control in an electric power steering |
GB2602476A (en) * | 2020-12-31 | 2022-07-06 | Zf Automotive Uk Ltd | Automotive vehicle lane keep assist system |
US11789412B2 (en) * | 2021-03-22 | 2023-10-17 | Steering Solutions Ip Holding Corporation | Functional limits for torque request based on neural network computing |
CN113978548B (en) * | 2021-11-12 | 2023-01-31 | 京东鲲鹏(江苏)科技有限公司 | Steering cooperative control method, device, equipment and medium applied to unmanned vehicle |
CN114235432B (en) * | 2021-11-12 | 2023-06-13 | 东风越野车有限公司 | Multi-source fusion diagnosis method and system for vehicle deviation problem cause |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101965286A (en) * | 2008-03-04 | 2011-02-02 | 日产自动车株式会社 | Lane keeping assist device and lane keeping assist method |
CN102112353A (en) * | 2008-07-29 | 2011-06-29 | 日产自动车株式会社 | Running control device, and running control method |
CN102700549A (en) * | 2011-02-18 | 2012-10-03 | 曼卡车和巴士股份公司 | Method for supporting drivers of vehicles, especially motor vehicles or commercial vehicles |
JP5469506B2 (en) * | 2010-03-30 | 2014-04-16 | 富士重工業株式会社 | Vehicle out-of-road departure prevention control device |
CN103832435A (en) * | 2012-11-20 | 2014-06-04 | 现代摩比斯株式会社 | Lane-keeping control device |
CN105059288A (en) * | 2015-08-11 | 2015-11-18 | 奇瑞汽车股份有限公司 | Lane keeping control system and method |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006264624A (en) * | 2005-03-25 | 2006-10-05 | Daimler Chrysler Ag | Lane maintaining assistant device |
CN101058319A (en) * | 2007-05-21 | 2007-10-24 | 林士云 | Electric assisting steering system based on intelligence control |
JP5389360B2 (en) * | 2008-01-09 | 2014-01-15 | 富士重工業株式会社 | Lane tracking control device and lane tracking control method |
US9542847B2 (en) * | 2011-02-16 | 2017-01-10 | Toyota Motor Engineering & Manufacturing North America, Inc. | Lane departure warning/assistance method and system having a threshold adjusted based on driver impairment determination using pupil size and driving patterns |
CN102616241A (en) * | 2012-03-28 | 2012-08-01 | 周圣砚 | Lane departure alarm system based on lane line model detection method and on-line study method |
CN102717825A (en) * | 2012-06-20 | 2012-10-10 | 清华大学 | Collaborative lane keeping controlling method |
CN106066644A (en) * | 2016-06-17 | 2016-11-02 | 百度在线网络技术(北京)有限公司 | Set up the method for intelligent vehicle control model, intelligent vehicle control method and device |
EP3266668A1 (en) * | 2016-07-06 | 2018-01-10 | Continental Automotive GmbH | Device for determining driving warning information |
CN107150682B (en) * | 2017-04-27 | 2019-08-02 | 同济大学 | A kind of lane holding auxiliary system |
CN107097785B (en) * | 2017-05-25 | 2019-08-27 | 江苏大学 | A kind of intelligent vehicle crosswise joint method that preview distance is adaptive |
CN107292048B (en) * | 2017-07-05 | 2020-12-04 | 合肥工业大学 | Lane keeping method and system based on veDYNA |
-
2018
- 2018-01-12 CN CN201810031566.9A patent/CN107972667B/en active Active
- 2018-01-12 CN CN201910298019.1A patent/CN109969181B/en active Active
- 2018-01-12 CN CN201910298010.0A patent/CN109969180B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101965286A (en) * | 2008-03-04 | 2011-02-02 | 日产自动车株式会社 | Lane keeping assist device and lane keeping assist method |
CN102112353A (en) * | 2008-07-29 | 2011-06-29 | 日产自动车株式会社 | Running control device, and running control method |
JP5469506B2 (en) * | 2010-03-30 | 2014-04-16 | 富士重工業株式会社 | Vehicle out-of-road departure prevention control device |
CN102700549A (en) * | 2011-02-18 | 2012-10-03 | 曼卡车和巴士股份公司 | Method for supporting drivers of vehicles, especially motor vehicles or commercial vehicles |
CN103832435A (en) * | 2012-11-20 | 2014-06-04 | 现代摩比斯株式会社 | Lane-keeping control device |
CN105059288A (en) * | 2015-08-11 | 2015-11-18 | 奇瑞汽车股份有限公司 | Lane keeping control system and method |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111158377A (en) * | 2020-01-15 | 2020-05-15 | 浙江吉利汽车研究院有限公司 | Transverse control method and system for vehicle and vehicle |
CN111158377B (en) * | 2020-01-15 | 2021-04-27 | 浙江吉利汽车研究院有限公司 | Transverse control method and system for vehicle and vehicle |
CN113120079A (en) * | 2020-01-15 | 2021-07-16 | 通用汽车环球科技运作有限责任公司 | Steering wheel angle offset correction for autonomous vehicles using angle control |
CN111175056A (en) * | 2020-01-17 | 2020-05-19 | 金龙联合汽车工业(苏州)有限公司 | Hardware-in-loop test device of commercial vehicle lane keeping system |
Also Published As
Publication number | Publication date |
---|---|
CN107972667B (en) | 2019-07-02 |
CN107972667A (en) | 2018-05-01 |
CN109969181B (en) | 2020-06-05 |
CN109969180B (en) | 2020-05-22 |
CN109969180A (en) | 2019-07-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109969181A (en) | A kind of deviation auxiliary system and its deviation householder method | |
CN108216231B (en) | One kind can open up united deviation auxiliary control method based on steering and braking | |
Wang et al. | Coordination control of differential drive assist steering and vehicle stability control for four-wheel-independent-drive EV | |
CN110187639B (en) | Trajectory planning control method based on parameter decision framework | |
Zhao et al. | Yaw and lateral stability control for four-wheel steer-by-wire system | |
Li et al. | Indirect shared control for cooperative driving between driver and automation in steer-by-wire vehicles | |
Zhang et al. | An analytical approach to improve vehicle maneuverability via torque vectoring control: theoretical study and experimental validation | |
Gao et al. | Robust lateral trajectory following control of unmanned vehicle based on model predictive control | |
CN109050661B (en) | Coordinated control method and coordinated control device for electronic differential and active differential steering | |
CN107719372B (en) | Four-drive electric car dynamics multi objective control system based on dynamic control allocation | |
Zhai et al. | MPC-based integrated control of trajectory tracking and handling stability for intelligent driving vehicle driven by four hub motor | |
Chen et al. | Dynamic drifting control for general path tracking of autonomous vehicles | |
CN110329255A (en) | A kind of deviation auxiliary control method based on man-machine coordination strategy | |
CN109017778A (en) | The expected path active steering control method of four motorized wheels vehicle | |
CN104462716B (en) | A kind of the brain-computer interface parameter and kinetic parameter design method of the brain control vehicle based on people's bus or train route model | |
CN109017760A (en) | Vehicle desired trajectory tracking, device and rolling time horizon optimization algorithm | |
Zhao et al. | Individual auxiliary and fault-tolerant control of steer-by-wire system considering different drivers steering characteristics | |
Lu et al. | A shared control design for steering assistance system considering driver behaviors | |
CN109291806B (en) | Lane departure auxiliary control system and control method of wheel hub motor driven automobile | |
Zhang et al. | Human-centered torque vectoring control for distributed drive electric vehicle considering driving characteristics | |
CN109017759A (en) | Expected path Vehicular yaw control method | |
Yu et al. | Nonlinear control of active four wheel steer-by-wire vehicles | |
CN109177744A (en) | A kind of torque distribution control method of deviation supplementary controlled system | |
CN109017446A (en) | Expect path vehicular longitudinal velocity tracing control method and device | |
Chen et al. | An improved IOHMM-based stochastic driver lane-changing model |
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 |