CN107792062A - Automatic parking control system - Google Patents
Automatic parking control system Download PDFInfo
- Publication number
- CN107792062A CN107792062A CN201710975324.0A CN201710975324A CN107792062A CN 107792062 A CN107792062 A CN 107792062A CN 201710975324 A CN201710975324 A CN 201710975324A CN 107792062 A CN107792062 A CN 107792062A
- Authority
- CN
- China
- Prior art keywords
- msub
- mrow
- mtd
- mtr
- msup
- 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 claims abstract description 14
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 238000004088 simulation Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims abstract description 6
- 230000008859 change Effects 0.000 claims description 6
- 230000005484 gravity Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 5
- 230000009897 systematic effect Effects 0.000 claims description 4
- 230000014509 gene expression Effects 0.000 claims description 3
- 238000003032 molecular docking Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 abstract description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000031068 symbiosis, encompassing mutualism through parasitism 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/06—Automatic manoeuvring for parking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
Abstract
The invention provides an automatic parking control system. The system divides the whole automatic parking process into an input layer, a strategy layer, a planning layer and a control layer. In the automatic parking control process, the input layer generates control instructions and parking track data sets of different types of vehicles through simulation; the strategy layer learns the simulation data by using a deep neural network algorithm and extracts the general relation between the control instruction and the parking track; the planning layer finds a proper parking strategy through training in a few steps, gives a control instruction under the parking scene and generates a planning track; and the control layer performs control feedback according to the deviation of the actual parking track and the planned track, so that the parking track is closest to the ideal track planned by the system.
Description
Technical field
The present invention relates to automatic parking field, more particularly to the automatic parking control system towards general scene.
Background technology
For many drivers, parallel parking is a kind of painful experience, and big city parking space is limited, by vapour
Car drives into narrow space turns into a required skill.Few situations for having stopped car without taking some twists and turns, parking may
Traffic jam, neurolysis and bumper is caused to be hit curved.With the development of automatic parking technology, above mentioned problem has obtained very big
Improvement.Automatic parking technology additionally aids some for solving densely populated city except that can help driver's automatic stopping
Parking and traffic problems.Sometimes, it can stop in small space and be limited by driver's technology.Automatic parking technology can be with
By automobile parking in less space, these spaces are more much smaller than the space that most of drivers oneself can stop.This just makes
Parking stall can be more easily found by obtaining car owner, while the space that the automobile of identical quantity takes is also smaller.
In the prior art, as publication number CN107102642A provides a kind of automatic parking system for pilotless automobile
System, it primarily focuses on automatic parking monitoring sensor-based system, and estimating for track of vehicle is carried out using the detection data of geomagnetic sensor
Calculate.Such as publication number CN106427996A offers a kind of multi-functional park control method and system, it is hindered by obtaining vehicle periphery
Hinder thing information;Automatic parking mode is selected according to vehicle periphery obstacle information or is remotely controlled mode of parking.Such as publication number CN
106043282 A provide a kind of full-automatic parking system and its control method for vehicle, and it is according to the periphery of the vehicle
Environmental information and the running state information of the vehicle cook up parking path, control electric boosting steering system and electronic stability
System and gear box control unit are completed automatically to park according to the parking path.
The related content of " automatic parking " is mainly that the hardware for laying particular emphasis on automated parking system forms, is each in the prior art
Part of module be how to work and each module between communication mode, its most of technology contents is without reference to parking scene
Consider, all have not been able to the parking problem for solving various parking scenes and different berth types.
The content of the invention
In order to solve the above-mentioned technical problem, the present invention provides a kind of automatic parking control system.
Specifically adopt the following technical scheme that:
The system includes input layer, strategic layer, planning layer and key-course;The input layer is used to receiving and perceiving current shape
The orientation and size of car speed, position and berth under state;The strategic layer is according to Current vehicle letter related to berth
Breath provides the control operation instruction of automatic parking;The kinetic model of the planning layer combination vehicle provides planning parking rail
Mark;The key-course is according to actual parking trajectory and plans that the deviation of parking trajectory carries out real-time feedback control.
Preferably, the input layer defines car according to car body size, position of centre of gravity, wheel base, front and back wheel rotary inertia
Type;Track Pick-up is carried out respectively to different type of vehicle:Whole docking process is divided into N number of stage, the control in N number of stage
Instruction set processed is δf N={ δf(1),…,δf(N)};Specific wagon control instruction comes from K pilot angle in each stage
Set Sδ={ δ1,…,δk, the vehicle of each type can obtain KNIndividual combined result;The time span in each stage is TN,
Total time span is T=T1+…+TN;Simulation step length is t, wherein TN=K*t;
The strategic layer learns to obtain vehicle according to vehicle final position, drift angle and speed by deep neural network
Control instruction combines;
The planning layer is according to control instruction set { δ1,δ2,…,δt, join with reference to current specific vehicle dynamic model
Number, draws vehicle parking track;
The key-course carries out feedback control, the final rail of parking of generation according to the deviation of actually park track and planned trajectory
Mark.
Preferably, the planning layer show that the concrete mode of vehicle parking track is:
1) in forward mode, according to vehicle-state equation of transfer
And
δf=δmax, δr=0,
Obtain
Wherein,For last moment Vehicular system state variable in forward mode,For subsequent time car in forward mode
System state variables;δfFor front wheel steering angle, δrFor rear-axle steering angle, δmaxFor steering locking angle;
2) in reversing mode, according to vehicle-state equation of transfer
And
δr=0, δf=δmax
Obtain
Wherein,For last moment Vehicular system state variable in reversing mode,For subsequent time in reversing mode
Vehicular system state variable.
Preferably, the key-course feedback control concretely comprises the following steps:
Feedback control is carried out according to the rotational angular velocity r under the angle β and inertial coodinate system of car speed and the vehicle longitudinal axis:
The state space vectors of original system are expressed as
The change real system of systematic parameter is
The state space vectors of reponse system are expressed as
Wherein, x=[β, r]TFor the state variable of idealized system;X '=[β ', r ']TFor the state variable of real system;A,
B is state constant;U is vehicle corner;K is feedback matrix;
By object functionMinimum, obtaining state feedback controller is
U=-Kx.
Preferably, what the key-course control was fed back concretely comprises the following steps:
Feedback control is carried out according to vehicle location and corner deviation:
By object function
Minimum, obtaining state feedback controller is
U=-Kx
Wherein:
(x ', y ') represents the coordinate of actual path point;
(x, y) represents the coordinate of ideal trajectory point;
The corner of ψ ' expressions actual path point;
ψ represents the corner of ideal trajectory point;
K=[k1, k2] represents feedback matrix.
The present invention has the advantages that:
(1) automated parking system proposed by the present invention is towards general scene, is looked for by Algorithm Learning and on-site training
To the parking strategy for adapting to current scene, solve the problems, such as that the applicable berth type of current shutdown system is single;
(2) two kinds of track of vehicle control modes based on control theory algorithm are proposed, than simply providing in the prior art
The mode of the correction value of one speed or deflection angle is more accurate.
Brief description of the drawings
Fig. 1 is automated parking system block schematic illustration.
Fig. 2 is two kinds of different automobile types simulation track schematic diagrames.
Fig. 3 is neural network structure schematic diagram.
Fig. 4 is planned trajectory schematic diagram.
Fig. 5 is vehicle dynamic model schematic diagram.
Fig. 6 is to carry out feedback control principle schematic diagram to δ _ f according to β and r.
Fig. 7 is that position and corner deviation carry out feedback control principle schematic diagram.
Embodiment
Whole automatic parking process is divided into four layers by the automatic parking control system towards general scene of the present invention:It is defeated
Enter layer, strategic layer, planning layer and key-course.As shown in Figure 1:
Input layer is used for receiving and perceiving car speed under current state, position and the orientation in berth and size;Plan
Slightly layer provides the control operation instruction of automatic parking according to the relevant information of Current vehicle and berth;Planning layer combination vehicle
Kinetic model provide planning parking trajectory;Key-course is according to actual parking trajectory and plans that the deviation of parking trajectory is carried out in fact
When feedback control.
The control instruction of different type vehicle is generated with parking by MATLAB simulation softwares and collateral learning algorithm first
Track data collection;Secondly emulation data are learnt using deep neural network algorithm, extracts control instruction and rail of parking
General relationship between mark, then when given one any one parking scene, training that can be Jing Guo a small number of steps is found
Suitable parking strategy, provide the control instruction under this parking lot scape;Then for the specific dynamics of vehicle of Current vehicle
Model, provide the track of parking under theoretical condition;Finally, feedback is controlled according to caused deviation during actually parking,
Make to park track closest to the ideal trajectory of systems organization.
(1) different vehicle type is defined:Different automobile types take on a different character parameter, such as car body size, position of centre of gravity, preceding
Rear axle is away from, front and back wheel rotary inertia etc., and we choose two kinds of vehicles, and A represents car, and B represents SUV.
The different vehicle type parameter example of table 1
(2) emulation generation control instruction and parking trajectory:Track is carried out respectively for two kinds of different type of vehicle A and B
Generation.Whole docking process is divided into N number of stage, the control instruction collection in N number of stage is combined into δf N={ δf(1),…,δf(N)};Often
Specific wagon control instruction comes from the set S of K pilot angle in the individual stageδ={ δ1,…,δk, therefore, each type
Vehicle can obtain KNIndividual combined result;The time span in each stage is TN, total time span is T=T1+…+TN;Emulation
Step-length is t, TN=K*t.Fig. 1 be two kinds of vehicle simulation track, δf N=-0.6, -0.4, -0.2,0,0.2,0.4,
0.6 }, K=7, N=4, t=0.01s, TN=3s, symbiosis is into 2401 tracks.
(3) neural network learning:In order in the case where learning vehicle final position, drift angle and speed, can determine one
Car is parked in final position by kind control instruction combination exactly to realize, we solve this using deep neural network algorithm
Problem.Fig. 2 is the structural representation of neutral net, is divided into input layer, hidden layer and output layer.Input as the position of vehicle target state
PutWith velocity v, the control instruction set { δ for needs is exported1,δ2,…,δt}。
(4) trajectory planning:In given control instruction set { δ1,δ2,…,δtOn the basis of, with reference to current specific vehicle
Kinetic parameters, provide vehicle parking track.Fig. 3 is the example of a planned trajectory.
Fig. 4 is vehicle dynamic model rough schematic view, there is as follows parameter used in model:
● v=car speeds
● the component of v_x=speed in the horizontal direction
● the component of v_y=speed in the vertical directions
● β=car speed and vehicle longitudinal axis angle
●Rotational angular velocity under=inertial coodinate system
● x=vehicle's center of gravity abscissas
● y=vehicle's center of gravity ordinates
● the side force of F=centers of gravity
●δf(δr)=front-wheel (trailing wheel) steering angle
● θ=vehicle front-wheel and berth long axis direction angle, direction of traffic conversion critical view angle
● ψ=from x-axis to the corner of vehicle major axis
1) forward mode
Vehicle-state equation of transfer is
WhereinFor last moment system state variables,For subsequent time system state variables.
Wherein:
Cf=μ cf,Cr=μ cr
Tire stiffness Coefficient mgV is to vehicle corner controlled quentity controlled variable Cf、CrInfluence coefficient;
Tire stiffness Coefficient mgV is to vehicle corner speed CflfInfluence coefficient;
Tyre rotation inertia IgzTo vehicle corner speed CflfInfluence coefficient;
Tyre rotation inertia is to vehicle corner speed CflfInfluence coefficient;
Influence coefficient of the front-wheel stiffness coefficient to front wheel angle controlled quentity controlled variable;
Influence coefficient of the trailing wheel stiffness coefficient to trailing wheel corner controlled quentity controlled variable;
Influence coefficient of the front tyre rotary inertia to front wheel angle controlled quentity controlled variable;
Influence coefficient of the rear tyre rotary inertia to trailing wheel corner controlled quentity controlled variable;
Cf=μ cf:Front tyre stiffness coefficient when coefficient of road adhesion is μ;
Cr=μ cr:Rear tyre stiffness coefficient when coefficient of road adhesion is μ;
(dry pavement μ=1, wet road surface μ=0.5).
During advance, δf=δmax, δr=0, then:
I.e.:
2) reversing mode
If the model is not influenceed by vehicle forerunner's rear-guard, reversing model can be regarded as regards headstock by the tailstock, still
Advance model so is used, then each parameter is exchanged before and after vehicle:
Wherein:
Therefore:
δ during reversingf *=δr=0, δr *=δf=δmax, therefore
Then:
No matter due to advancing or retreating, β and r are incremental form, so defining identical;But ψ is cumulant, currently
When entering,
ψ*=ψ+r* Δs t;
When reversing, the tailstock is considered as headstock, so
ψ*=ψ+r* Δ t+ π;
Velocity component
v_x*=| v | * cos (β+ψ*);
v_y*=| v | * sin (β+ψ*);
Position coordinates
x*=x+vx*Δt;
y*=y+vy*Δt;
Deflection angle
ψ*=ψ+r* Δs t;(advance)
ψ*=ψ+r* Δ t+ π;(reversing).
(5) feedback control:Due to reasons such as vehicle abrasion, environmental changes, the model parameter in actual vehicle operating system
It may be had differences with the model parameter of preferable planning system, it is inclined between actual parking trajectory and planned trajectory so as to cause
Difference.In order to eliminate as much as or reduce this deviation, we carry out feedback control using linear control method to system.
In actual shutdown system, due to the intrinsic parameter of vehicle such as mg、lf(lr)、Cf(Cr) difference, vehicle running orbit meeting
Had differences with the track under idealized system.To eliminate this species diversity, negative feedback control is added to system, to adjust vehicle operation
Track.
The feedback control of track of vehicle regulation is divided into two kinds:First, according to internal system variable quantity β and r deviation to δfEnter
Row feedback control;Second, it is position and angular deviation to δ according to system cumulantfCarry out feedback control.
1) according to β and r to δfCarry out feedback control
It can be seen from vehicle dynamic model, the change of systematic parameter can cause system change amount β and r change, and then
Influence track of vehicle.Feedback principle is as shown in Figure 5.
The state space vectors of original system can be expressed as
Because the change real system of systematic parameter is
According to LQR control methods, we will design a state feedback controller u=-Kx and cause object functionIt is minimum.So, the state space vectors of reponse system can be expressed as
Wherein x=[β, r]T:The state variable of idealized system;
X '=[β ', r ']T:The state variable of real system;
A, B are state constant;
U=δf:Vehicle corner;
K:Feedback matrix K=[k1, k2], circular refer to LQR control algolithms.
Emulation step number is set to 12000 steps in MATLAB discrete systems, step-length 0.001s, is divided into four-stage:
stage1:Step1-step3000, δf=-0.4 (v=-1)/δf=0.4 (v=1);
stage2:Step3001-step6000, δf=0;
stage3:Step6001-step9000, δf=0.4 (v=-1)/δf=-0.4 (v=1);
stage4:Step9001-step12000, δf=0.Emulation step number is set to 12000 in MATLAB discrete systems
Step, step-length 0.001s, is divided into four-stage:
stage1:Step1-step3000, δf=-0.4 (v=-1)/δf=0.4 (v=1);
stage2:Step3001-step6000, δf=0;
stage3:Step6001-step9000, δf=0.4 (v=-1)/δf=-0.4 (v=1);
stage4:Step9001-step12000, δf=0.
2) feedback control is carried out according to vehicle location and corner deviation, feedback principle is as shown in Figure 6.
On the premise of knowing last moment feedback ideal trajectory point with actual path point, we can find a steering
Angle causes the actual path of vehicle next step closest to the track of path planning, and causes object function
It is minimum.Wherein:
(x ', y ') represents the coordinate of actual path point;
(x, y) represents the coordinate of ideal trajectory point;
The corner of ψ ' expressions actual path point;
ψ represents the corner of ideal trajectory point;
K=[k1, k2] represents feedback matrix, herein k1, k2=1;
According to the limitation of vehicle steering locking angle, δ is takenf∈[-0.70,0.70].Feedback procedure is:
●step1:Original system δf=0.4, reponse system δf=0.4
●step2-step300:Original system δf=0.4, [- 0.70,0.70]
●step301-step600:Original system δf=0, reponse system δf=[- 0.70,0.70]
●step601-step900:Original system δf=-0.4, reponse system δf=[- 0.70,0.70]
●step900-step1200:Original system δf=0, reponse system δf=[- 0.70,0.70].
Claims (5)
- A kind of 1. automatic parking control system, it is characterised in thatThe system includes input layer, strategic layer, planning layer and key-course;Car speed, position and the orientation in berth and the size that the input layer is used to receive and perceive under current state;The control operation that the strategic layer provides automatic parking according to the relevant information of Current vehicle and berth instructs;The kinetic model of the planning layer combination vehicle provides planning parking trajectory;The key-course is according to actual parking trajectory and plans that the deviation of parking trajectory carries out real-time feedback control.
- A kind of 2. automatic parking control system as claimed in claim 2, it is characterised in thatThe input layer defines type of vehicle according to car body size, position of centre of gravity, wheel base, front and back wheel rotary inertia;To not Same type of vehicle carries out Track Pick-up respectively:Whole docking process is divided into N number of stage, the control instruction set in N number of stage For δf N={ δf(1),…,δf(N)};Specific wagon control instruction comes from the set S of K pilot angle in each stageδ= {δ1,…,δk, the vehicle of each type can obtain KNIndividual combined result;The time span in each stage is TN, total time Length is T=T1+…+TN;Simulation step length is t, wherein TN=K*t;The strategic layer learns to obtain the control of vehicle according to vehicle final position, drift angle and speed by deep neural network Instructing combination;The planning layer is according to control instruction set { δ1,δ2,…,δt, with reference to current specific vehicle dynamic model parameter, Draw vehicle parking track;The key-course carries out feedback control, the final track of parking of generation according to the deviation of actually park track and planned trajectory.
- 3. a kind of automatic parking control method towards general parking scene as claimed in claim 2, it is characterised in that described Planning layer show that the concrete mode of vehicle parking track is:1) in forward mode, according to vehicle-state equation of transfer<mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mover> <mi>&beta;</mi> <mo>&CenterDot;</mo> </mover> </mtd> </mtr> <mtr> <mtd> <mover> <mi>r</mi> <mo>&CenterDot;</mo> </mover> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>a</mi> <mn>11</mn> </msub> </mtd> <mtd> <msub> <mi>a</mi> <mn>12</mn> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>a</mi> <mn>21</mn> </msub> </mtd> <mtd> <msub> <mi>a</mi> <mn>22</mn> </msub> </mtd> </mtr> </mtable> </mfenced> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mi>&beta;</mi> </mtd> </mtr> <mtr> <mtd> <mi>r</mi> </mtd> </mtr> </mtable> </mfenced> <mo>+</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>b</mi> <mn>11</mn> </msub> </mtd> <mtd> <msub> <mi>b</mi> <mn>12</mn> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>b</mi> <mn>21</mn> </msub> </mtd> <mtd> <msub> <mi>b</mi> <mn>22</mn> </msub> </mtd> </mtr> </mtable> </mfenced> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>&delta;</mi> <mi>f</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>&delta;</mi> <mi>r</mi> </msub> </mtd> </mtr> </mtable> </mfenced> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>a</mi> <mn>11</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mrow> <msub> <mi>c</mi> <mi>f</mi> </msub> <mo>+</mo> <msub> <mi>c</mi> <mi>r</mi> </msub> </mrow> <mrow> <msub> <mi>m</mi> <mi>g</mi> </msub> <mi>v</mi> </mrow> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>a</mi> <mn>12</mn> </msub> <mo>=</mo> <mo>-</mo> <mn>1</mn> <mo>-</mo> <mfrac> <mrow> <msub> <mi>c</mi> <mi>f</mi> </msub> <msub> <mi>l</mi> <mi>f</mi> </msub> <mo>-</mo> <msub> <mi>c</mi> <mi>r</mi> </msub> <msub> <mi>l</mi> <mi>r</mi> </msub> </mrow> <mrow> <msub> <mi>m</mi> <mi>g</mi> </msub> <msup> <mi>v</mi> <mn>2</mn> </msup> </mrow> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>a</mi> <mn>21</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mrow> <msub> <mi>c</mi> <mi>f</mi> </msub> <msub> <mi>l</mi> <mi>f</mi> </msub> <mo>-</mo> <msub> <mi>c</mi> <mi>r</mi> </msub> <msub> <mi>l</mi> <mi>r</mi> </msub> </mrow> <msub> <mi>I</mi> <mrow> <mi>g</mi> <mi>z</mi> </mrow> </msub> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>a</mi> <mn>22</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mrow> <msub> <mi>c</mi> <mi>f</mi> </msub> <msup> <msub> <mi>l</mi> <mi>f</mi> </msub> <mn>2</mn> </msup> <mo>+</mo> <msub> <mi>c</mi> <mi>r</mi> </msub> <msup> <msub> <mi>l</mi> <mi>r</mi> </msub> <mn>2</mn> </msup> </mrow> <mrow> <msub> <mi>I</mi> <mrow> <mi>g</mi> <mi>z</mi> </mrow> </msub> <mi>v</mi> </mrow> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>b</mi> <mn>11</mn> </msub> <mo>=</mo> <mfrac> <msub> <mi>c</mi> <mi>f</mi> </msub> <mrow> <msub> <mi>m</mi> <mi>g</mi> </msub> <mi>v</mi> </mrow> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>b</mi> <mn>21</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>c</mi> <mi>f</mi> </msub> <msub> <mi>l</mi> <mi>f</mi> </msub> </mrow> <msub> <mi>I</mi> <mrow> <mi>g</mi> <mi>z</mi> </mrow> </msub> </mfrac> </mrow> </mtd> </mtr> </mtable> </mfenced>Andδf=δmax, δr=0,Obtain<mrow> <mover> <mi>&beta;</mi> <mo>&CenterDot;</mo> </mover> <mo>=</mo> <msub> <mi>a</mi> <mn>11</mn> </msub> <mi>&beta;</mi> <mo>+</mo> <msub> <mi>a</mi> <mn>12</mn> </msub> <mi>r</mi> <mo>+</mo> <msub> <mi>b</mi> <mn>11</mn> </msub> <msub> <mi>&delta;</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </mrow><mrow> <mover> <mi>r</mi> <mo>&CenterDot;</mo> </mover> <mo>=</mo> <msub> <mi>a</mi> <mn>21</mn> </msub> <mi>&beta;</mi> <mo>+</mo> <msub> <mi>a</mi> <mn>22</mn> </msub> <mi>r</mi> <mo>+</mo> <msub> <mi>b</mi> <mn>21</mn> </msub> <msub> <mi>&delta;</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </mrow>Wherein,For last moment Vehicular system state variable in forward mode,For subsequent time vehicle system in forward mode System state variable;δfFor front wheel steering angle, δrFor rear-axle steering angle, δmaxFor steering locking angle;2) in reversing mode, according to vehicle-state equation of transfer<mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msup> <mover> <mi>&beta;</mi> <mo>&CenterDot;</mo> </mover> <mo>*</mo> </msup> </mtd> </mtr> <mtr> <mtd> <msup> <mover> <mi>r</mi> <mo>&CenterDot;</mo> </mover> <mo>*</mo> </msup> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>a</mi> <mn>11</mn> </msub> </mtd> <mtd> <mrow> <mo>-</mo> <mn>2</mn> <mo>-</mo> <msub> <mi>a</mi> <mn>12</mn> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>-</mo> <msub> <mi>a</mi> <mn>21</mn> </msub> </mrow> </mtd> <mtd> <msub> <mi>a</mi> <mn>22</mn> </msub> </mtd> </mtr> </mtable> </mfenced> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msup> <mi>&beta;</mi> <mo>*</mo> </msup> </mtd> </mtr> <mtr> <mtd> <msup> <mi>r</mi> <mo>*</mo> </msup> </mtd> </mtr> </mtable> </mfenced> <mo>+</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>b</mi> <mn>12</mn> </msub> </mtd> <mtd> <msub> <mi>b</mi> <mn>11</mn> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>b</mi> <mn>22</mn> </msub> </mtd> <mtd> <msub> <mi>b</mi> <mn>21</mn> </msub> </mtd> </mtr> </mtable> </mfenced> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msup> <msub> <mi>&delta;</mi> <mi>f</mi> </msub> <mo>*</mo> </msup> </mtd> </mtr> <mtr> <mtd> <msup> <msub> <mi>&delta;</mi> <mi>r</mi> </msub> <mo>*</mo> </msup> </mtd> </mtr> </mtable> </mfenced> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>b</mi> <mn>12</mn> </msub> <mo>=</mo> <mfrac> <msub> <mi>C</mi> <mi>r</mi> </msub> <mrow> <msub> <mi>m</mi> <mi>g</mi> </msub> <mi>v</mi> </mrow> </mfrac> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>b</mi> <mn>22</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <mrow> <msub> <mi>C</mi> <mi>r</mi> </msub> <msub> <mi>l</mi> <mi>r</mi> </msub> </mrow> <msub> <mi>I</mi> <mrow> <mi>g</mi> <mi>z</mi> </mrow> </msub> </mfrac> </mrow> </mtd> </mtr> </mtable> </mfenced>Andδr=0, δf=δmaxObtain<mrow> <msup> <mover> <mi>&beta;</mi> <mo>&CenterDot;</mo> </mover> <mo>*</mo> </msup> <mo>=</mo> <msub> <mi>a</mi> <mn>11</mn> </msub> <msup> <mi>&beta;</mi> <mo>*</mo> </msup> <mo>+</mo> <mrow> <mo>(</mo> <mo>-</mo> <mn>2</mn> <mo>-</mo> <msub> <mi>a</mi> <mn>12</mn> </msub> <mo>)</mo> </mrow> <msup> <mi>r</mi> <mo>*</mo> </msup> <mo>+</mo> <msub> <mi>b</mi> <mn>11</mn> </msub> <msub> <mi>&delta;</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </mrow><mrow> <msup> <mover> <mi>r</mi> <mo>&CenterDot;</mo> </mover> <mo>*</mo> </msup> <mo>=</mo> <mo>-</mo> <msub> <mi>a</mi> <mn>21</mn> </msub> <msup> <mi>&beta;</mi> <mo>*</mo> </msup> <mo>+</mo> <msub> <mi>a</mi> <mn>22</mn> </msub> <msup> <mi>r</mi> <mo>*</mo> </msup> <mo>+</mo> <msub> <mi>b</mi> <mn>21</mn> </msub> <msub> <mi>&delta;</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> </mrow>Wherein,For last moment Vehicular system state variable in reversing mode,For subsequent time vehicle in reversing mode System state variables, a11For tire stiffness Coefficient mgInfluence coefficients of the v to vehicle corner controlled quentity controlled variable;a12For tire stiffness coefficient mgV is to vehicle corner speed CflfInfluence coefficient;a21For tyre rotation inertia IgzTo the influence coefficient of vehicle corner speed;a22 Influence coefficient for tyre rotation inertia to vehicle corner speed;b11For influence of the front-wheel stiffness coefficient to front wheel angle controlled quentity controlled variable Coefficient;b12Influence coefficient for trailing wheel stiffness coefficient to trailing wheel corner controlled quentity controlled variable;b21It is front tyre rotary inertia to preceding rotation The influence coefficient of angle controlled quentity controlled variable;b22Influence coefficient for rear tyre rotary inertia to trailing wheel corner controlled quentity controlled variable;CfIt is attached for road surface Front tyre stiffness coefficient when coefficient is μ;CrRear tyre stiffness coefficient when for coefficient of road adhesion being μ.
- 4. a kind of automatic parking control method towards general parking scene as claimed in claim 2, it is characterised in that described Key-course feedback control concretely comprises the following steps:Feedback control is carried out according to the rotational angular velocity r under the angle β and inertial coodinate system of car speed and the vehicle longitudinal axis:The state space vectors of original system are expressed as<mrow> <mover> <mi>x</mi> <mo>&CenterDot;</mo> </mover> <mo>=</mo> <mi>A</mi> <mi>x</mi> <mo>+</mo> <mi>B</mi> <mi>u</mi> </mrow>The change real system of systematic parameter is<mrow> <msup> <mover> <mi>x</mi> <mo>&CenterDot;</mo> </mover> <mo>&prime;</mo> </msup> <mo>=</mo> <msup> <mi>A</mi> <mo>&prime;</mo> </msup> <msup> <mi>x</mi> <mo>&prime;</mo> </msup> <mo>+</mo> <msup> <mi>B</mi> <mo>&prime;</mo> </msup> <mi>u</mi> </mrow>The state space vectors of reponse system are expressed as<mrow> <msup> <mi>x</mi> <mo>&prime;</mo> </msup> <mo>=</mo> <msup> <mi>A</mi> <mo>&prime;</mo> </msup> <msup> <mi>x</mi> <mo>&prime;</mo> </msup> <mo>+</mo> <msup> <mi>B</mi> <mo>&prime;</mo> </msup> <mover> <mo>&lsqb;</mo> <mo>&CenterDot;</mo> </mover> <mi>u</mi> <mo>-</mo> <mi>K</mi> <mrow> <mo>(</mo> <msup> <mi>x</mi> <mo>&prime;</mo> </msup> <mo>-</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> </mrow>Wherein, x=[β, r]TFor the state variable of idealized system;X '=[β ', r ']TFor the state variable of real system;A, B are State constant;U is vehicle corner;K is feedback matrix;By object functionMinimum, obtaining state feedback controller isU=-Kx.
- 5. a kind of automatic parking control method towards general parking scene as claimed in claim 2, it is characterised in that described Key-course control feedback concretely comprises the following steps:Feedback control is carried out according to vehicle location and corner deviation:By object function<mrow> <mi>d</mi> <mi>i</mi> <mi>s</mi> <mo>=</mo> <mi>k</mi> <mn>1</mn> <msqrt> <mrow> <msup> <mrow> <mo>(</mo> <msup> <mi>x</mi> <mo>&prime;</mo> </msup> <mo>-</mo> <mi>x</mi> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>+</mo> <msup> <mrow> <mo>(</mo> <msup> <mi>y</mi> <mo>&prime;</mo> </msup> <mo>-</mo> <mi>y</mi> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> <mo>+</mo> <mi>k</mi> <mn>2</mn> <mo>|</mo> <msup> <mi>&psi;</mi> <mo>&prime;</mo> </msup> <mo>-</mo> <mi>&psi;</mi> <mo>|</mo> </mrow> Minimum, obtaining state feedback controller isU=-Kx is wherein:(x ', y ') represents the coordinate of actual path point;(x, y) represents the coordinate of ideal trajectory point;The corner of ψ ' expressions actual path point;ψ represents the corner of ideal trajectory point;K=[k1, k2] represents feedback matrix.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710975324.0A CN107792062B (en) | 2017-10-16 | 2017-10-16 | Automatic parking control system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710975324.0A CN107792062B (en) | 2017-10-16 | 2017-10-16 | Automatic parking control system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107792062A true CN107792062A (en) | 2018-03-13 |
CN107792062B CN107792062B (en) | 2019-11-05 |
Family
ID=61534130
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710975324.0A Active CN107792062B (en) | 2017-10-16 | 2017-10-16 | Automatic parking control system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107792062B (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108407805A (en) * | 2018-03-30 | 2018-08-17 | 中南大学 | A kind of vehicle automatic parking method based on DQN |
CN108860139A (en) * | 2018-04-11 | 2018-11-23 | 浙江零跑科技有限公司 | A kind of automatic parking method for planning track based on depth enhancing study |
CN109823348A (en) * | 2019-02-19 | 2019-05-31 | 百度在线网络技术(北京)有限公司 | Unmanned vehicle reversing model control method, device, equipment and computer-readable medium |
CN110525428A (en) * | 2019-08-29 | 2019-12-03 | 合肥工业大学 | A kind of automatic parking method based on the study of fuzzy deeply |
WO2020001423A1 (en) * | 2018-06-29 | 2020-01-02 | 华为技术有限公司 | Method and device for determining automatic parking strategy |
CN110654373A (en) * | 2018-06-29 | 2020-01-07 | 比亚迪股份有限公司 | Automatic parking method and device and vehicle |
CN110654374A (en) * | 2018-06-29 | 2020-01-07 | 比亚迪股份有限公司 | Automatic parking simulation system and test method thereof |
WO2020056875A1 (en) * | 2018-09-20 | 2020-03-26 | 初速度(苏州)科技有限公司 | Parking strategy based on deep reinforcement learning |
CN111047948A (en) * | 2019-11-27 | 2020-04-21 | 康忠文 | Auxiliary method and system for learning to learn to train |
CN111098852A (en) * | 2019-12-02 | 2020-05-05 | 北京交通大学 | Parking path planning method based on reinforcement learning |
CN111098850A (en) * | 2018-10-25 | 2020-05-05 | 北京初速度科技有限公司 | Automatic parking auxiliary system and automatic parking method |
CN111434550A (en) * | 2019-01-15 | 2020-07-21 | 初速度(苏州)科技有限公司 | Simulation-based parking strategy generation method and system |
CN113110500A (en) * | 2021-05-10 | 2021-07-13 | 的卢技术有限公司 | Transverse control method for low-speed automatic parking of automatic-driving automobile |
CN113252366A (en) * | 2021-06-23 | 2021-08-13 | 奇瑞新能源汽车股份有限公司 | Automatic parking evaluation method and device for vehicle, vehicle and storage medium |
CN113978547A (en) * | 2021-10-21 | 2022-01-28 | 江铃汽车股份有限公司 | Automatic driving steering control method and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104057950A (en) * | 2014-05-21 | 2014-09-24 | 重庆长安汽车股份有限公司 | Full-automatic parking system |
CN104260722A (en) * | 2014-09-23 | 2015-01-07 | 北京理工大学 | Automatic parking system |
CN105857306A (en) * | 2016-04-14 | 2016-08-17 | 中国科学院合肥物质科学研究院 | Vehicle autonomous parking path programming method used for multiple parking scenes |
US20170294121A1 (en) * | 2016-04-12 | 2017-10-12 | Ford Global Technologies, Llc | Detecting available parking spaces |
-
2017
- 2017-10-16 CN CN201710975324.0A patent/CN107792062B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104057950A (en) * | 2014-05-21 | 2014-09-24 | 重庆长安汽车股份有限公司 | Full-automatic parking system |
CN104260722A (en) * | 2014-09-23 | 2015-01-07 | 北京理工大学 | Automatic parking system |
US20170294121A1 (en) * | 2016-04-12 | 2017-10-12 | Ford Global Technologies, Llc | Detecting available parking spaces |
CN105857306A (en) * | 2016-04-14 | 2016-08-17 | 中国科学院合肥物质科学研究院 | Vehicle autonomous parking path programming method used for multiple parking scenes |
Non-Patent Citations (2)
Title |
---|
吴冰等: "基于RBF神经网络的自动泊车路径规划", 《合肥工业大学学报(自然科学版)》 * |
张放等: "基于模糊逻辑的自动平行泊车控制方法", 《汽车工程》 * |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108407805A (en) * | 2018-03-30 | 2018-08-17 | 中南大学 | A kind of vehicle automatic parking method based on DQN |
CN108860139A (en) * | 2018-04-11 | 2018-11-23 | 浙江零跑科技有限公司 | A kind of automatic parking method for planning track based on depth enhancing study |
CN108860139B (en) * | 2018-04-11 | 2019-11-29 | 浙江零跑科技有限公司 | A kind of automatic parking method for planning track based on depth enhancing study |
CN110654374B (en) * | 2018-06-29 | 2022-03-18 | 比亚迪股份有限公司 | Automatic parking simulation system and test method thereof |
US11897454B2 (en) | 2018-06-29 | 2024-02-13 | Huawei Technologies Co., Ltd. | Method and apparatus for determining automatic parking strategy |
WO2020001423A1 (en) * | 2018-06-29 | 2020-01-02 | 华为技术有限公司 | Method and device for determining automatic parking strategy |
CN110654373A (en) * | 2018-06-29 | 2020-01-07 | 比亚迪股份有限公司 | Automatic parking method and device and vehicle |
CN110654374A (en) * | 2018-06-29 | 2020-01-07 | 比亚迪股份有限公司 | Automatic parking simulation system and test method thereof |
WO2020056875A1 (en) * | 2018-09-20 | 2020-03-26 | 初速度(苏州)科技有限公司 | Parking strategy based on deep reinforcement learning |
CN111098850A (en) * | 2018-10-25 | 2020-05-05 | 北京初速度科技有限公司 | Automatic parking auxiliary system and automatic parking method |
CN111434550A (en) * | 2019-01-15 | 2020-07-21 | 初速度(苏州)科技有限公司 | Simulation-based parking strategy generation method and system |
CN111434550B (en) * | 2019-01-15 | 2021-12-31 | 魔门塔(苏州)科技有限公司 | Simulation-based parking strategy generation method and system |
CN109823348A (en) * | 2019-02-19 | 2019-05-31 | 百度在线网络技术(北京)有限公司 | Unmanned vehicle reversing model control method, device, equipment and computer-readable medium |
CN110525428B (en) * | 2019-08-29 | 2020-09-04 | 合肥工业大学 | Automatic parking method based on fuzzy depth reinforcement learning |
CN110525428A (en) * | 2019-08-29 | 2019-12-03 | 合肥工业大学 | A kind of automatic parking method based on the study of fuzzy deeply |
CN111047948A (en) * | 2019-11-27 | 2020-04-21 | 康忠文 | Auxiliary method and system for learning to learn to train |
CN111098852A (en) * | 2019-12-02 | 2020-05-05 | 北京交通大学 | Parking path planning method based on reinforcement learning |
CN113110500A (en) * | 2021-05-10 | 2021-07-13 | 的卢技术有限公司 | Transverse control method for low-speed automatic parking of automatic-driving automobile |
CN113252366A (en) * | 2021-06-23 | 2021-08-13 | 奇瑞新能源汽车股份有限公司 | Automatic parking evaluation method and device for vehicle, vehicle and storage medium |
CN113978547A (en) * | 2021-10-21 | 2022-01-28 | 江铃汽车股份有限公司 | Automatic driving steering control method and system |
CN113978547B (en) * | 2021-10-21 | 2022-08-02 | 江铃汽车股份有限公司 | Automatic driving steering control method and system |
Also Published As
Publication number | Publication date |
---|---|
CN107792062B (en) | 2019-11-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107792062B (en) | Automatic parking control system | |
CN107697065B (en) | Automatic parking control method for general parking scene | |
Polack et al. | The kinematic bicycle model: A consistent model for planning feasible trajectories for autonomous vehicles? | |
CN110235071B (en) | Automatic car guiding and trajectory tracking | |
Marzbani et al. | Autonomous vehicles: Autodriver algorithm and vehicle dynamics | |
Li et al. | A potential field approach-based trajectory control for autonomous electric vehicles with in-wheel motors | |
CN103121451B (en) | A kind of detour changes the tracking and controlling method of track | |
CN103057436B (en) | Yawing moment control method of individual driven electromobile based on multi-agent | |
Velenis et al. | Modeling aggressive maneuvers on loose surfaces: The cases of trail-braking and pendulum-turn | |
CN105857306A (en) | Vehicle autonomous parking path programming method used for multiple parking scenes | |
CN107618504A (en) | It is a kind of applied to the crawl speed control method and device automatically parked | |
Subosits et al. | Impacts of model fidelity on trajectory optimization for autonomous vehicles in extreme maneuvers | |
CN106926840A (en) | A kind of vehicle extremal dynamic model Trajectory Tracking Control System | |
US20220169247A1 (en) | Motor-vehicle trajectory planning and control to cause automated motor-vehicles to perform low-speed manoeuvres in automated driving | |
Subosits et al. | Autonomous vehicle control for emergency maneuvers: The effect of topography | |
CN107490968A (en) | The adaptive layered of autonomous driving vehicle passs rank path tracking control method | |
DE102019124318A1 (en) | METHOD AND SYSTEM FOR OPERATING A VEHICLE | |
Lin et al. | Coordinated control architecture for motion management in ADAS systems | |
Menhour et al. | Multivariable decoupled longitudinal and lateral vehicle control: A model-free design | |
De Castro et al. | Minimum-time path-following for highly redundant electric vehicles | |
Kang et al. | Design and testing of a controller for autonomous vehicle path tracking using GPS/INS sensors | |
Chen et al. | Car-like ground vehicle trajectory tracking by using trajectory linearization control | |
CN111444577A (en) | Automatic avoidance method for electric motor coach | |
CN207274660U (en) | It is a kind of applied to the crawl speed control device automatically parked | |
CN115268464A (en) | Automatic driving automobile track tracking control method and device based on hierarchical constraint |
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 |