CN201235814Y - Electric power-assisted steering apparatus - Google Patents

Abstract

The utility model relates to an electrical power steering gear which comprises a torque sensor, a vehicle speed sensor, an electric current sensor, a control unit ECU, a power electric motor, a retarding mechanism and a wheel stabling controller that consists of a yaw velocity instruction generator, a state observer, a torque apportioning controller and a tyre skidding controller. The electrical power steering gear can solve the problems of understeer and oversteer caused by a road surface and a vehicle condition to keep the electrical power system steering accurately, or an EPS device and the control algorithm of the EPS device which can still guarantee the normal steer of the automobile and have the failure security control function when a steering power system does not work and the steering power function can not be kept. The torques of various wheels are adjusted to form the car body steering torque under the circumstance that the prior EPS system does not work or the understeer and oversteer happen so as to guarantee the completion of the steering action, realize the EPS failure security control and avoid a traffic accident. The utility model has the advantages of quick operation, reliable operation, low cost, and the like.

Description

Electric power-assisted steering apparatus

Technical field

The utility model relates to a kind of electric power-assisted steering apparatus, is specially the fail-safe-control strategy that a kind of four-wheel braking distributes the electric power-assisted steering apparatus of control.

Background technology

Electric boosting steering system (EPS, Electric Power Steering) is formed as the key of vehicle active safety, safety and road-holding property when it directly affects vehicle 2 operations.Along with the fast development of modern automobile technology, the safety of automotive steering structure also becomes the emphasis of research gradually.In recent years, along with people are more and more higher to safety, environmental protection and energy-conservation cry, automobile electric booster steering system is subjected to the generally favor of industry as the steering hardware of a kind of " type as required ", has represented the developing direction of following motor turning technology.The defective that exists in the current techniques has: (1) EPS itself is difficult to compensate understeer (under steering) and ovdersteering (over steering) problem that causes owing to road surface and vehicle condition reason, (2) under the situation that EPS lost efficacy, self can't keep the power steering controllable function.When vehicle 2 quickly travels, the adhesive ability of asking on wheel and road surface reduces, even best chaufeur also is difficult to the automobile of running at high speed is remained on the predetermined route, automobile takes place to break away and sideslip easily, lose directional balance, even roll-over accident takes place in sharply turning, at this moment just need of the generation of EPS system to avoid traffic accident.But when the EPS system breaks down can not normally move the time, roll-over accident takes place easily when turning.

The Security Control Problem of the EPS system failure can be expressed as follows: in vehicle traveling process, because of external interference, such as the pedestrian, vehicle 2 or environment etc. change (icing suddenly, wet and slippery, and rubble etc.) under the situation, chaufeur takes some promptly to dodge measure, make automobile enter unstable motoring condition, precarious positions such as predetermined running route or upset trend promptly appear departing from, and this moment the EPS system any one or several links et out of order, chaufeur is made a response, when bearing circle 9 has angle, can utilize steering wheel angle sensor to detect the angle that bearing circle 9 produces, adopt the electricity consumption of X-By-Wire system to control and to reduce widely to postpone,, can utilize torque sensor to detect during bearing circle 9 non-angular to twist moment to reach control vehicle body for the emergency treatment dangerous situation has under won valuable time.Thereby can be automatically and help chaufeur to improve stability of automobile in time.

Summary of the invention

The purpose of this utility model is to provide a kind of electric power-assisted steering apparatus, can overcome the defective of prior art.The utility model is to four-wheel braking moment or drive torque signal allocation, auxiliary EPS turn to or the car body of calming dynamic, guarantee finishing of handling maneuver, realize the EPS fail-safe-control, guarantee the safety control of vehicle.

The electric power-assisted steering apparatus that the utility model provides comprises: torque sensor, car speed sensor, current sensor, control unit ECU (Electronic Control Unit), assisted electric machine and speed reduction gearing; Bearing circle is linked together by input shaft and torque sensor, and torque sensor also links together with the assisted electric machine, and torque sensor all is connected with control unit ECU with the assisted electric machine.Also comprise the wheel stability controller, consist of:

Wheel stability controller and four wheel drive or lock torque control signal output unit, the consisting of of wheel stability controller:

The yaw velocity command generator is used to obtain the yaw velocity command value;

State observer is used to obtain actual yaw velocity value and sideslip angle value;

The torque distribution controller is added to different big or small driving torque or braking torques respectively on each wheel, forms steering torque, and steering response is speeded up;

The tyre slip controller by regulating the application force between vehicle tyre and road surface, improves the active safety performance of automobile.

Described yaw velocity command generator is by detecting the steering angle δ of wheel flutter _fAnd the difference of angle of side slip β multiply by speed v again divided by between center of gravity and the wheel flutter apart from l _fObtain yaw velocity command value γ *.

Described state observer comprises two inputs, and promptly u and y are output as

, observer contains n integrator and whole state variables is made an estimate, and G is an observer output feedback battle array, its handle

Inverse feedback extremely

The place makes to configuration observer limit, improves its dynamic property, makes as early as possible

Approach zero and introduce, it is a kind of output feedback.

Described torque distribution controller is to stablize needed moment of rotation size according to car body, front and back wheel group mean allocation drive torque, left and right wheels group complementary allocated steering torque; Perhaps only adopt and on front-wheel or trailing wheel, torque is averaged distribution.

Described tyre slip controller is based on wheel velocity and calculates acceleration/accel and propulsive effort signal, and observation car body equivalent moment of inertia makes up external disturbance signal observer, thereby forms the tyre slip controller.

The step that the electric power-assisted steering control method that the utility model provides comprises:

1) by corner that detects bearing circle or the steering angle that torque obtains wheel flutter;

2) the yaw velocity command value that obtains expecting by vehicle body yaw velocity command generator;

3) obtain actual yaw velocity and sideslip angle value by state observer;

4) adopt feedforward and pi regulator to regulate the assisted diversion torque, be added to respectively on each wheel, form steering torque, make steering response speed up, move accurately by the driving torque or the braking torque of torque distribution controller with different sizes.

5) the tyre slip controller improves the active safety performance of automobile by regulating the application force between vehicle tyre and road surface.

When EPS lost efficacy, determine the steering angle sigma that chaufeur is wished by the corner or the torque that detect bearing circle _f, regulate the driving torque or the braking torque of each wheel, form the vehicle body steering torque, force vehicle body to require to turn to according to the control of chaufeur, realize the fail safe function of EPS.

The utility model provides a kind of electric power-assisted steering apparatus and control method thereof can overcome the defective of prior art.The utility model is to four-wheel braking and drive torque signal allocation, auxiliary EPS turn to or the car body of calming dynamic, guarantee finishing of handling maneuver, realize the EPS fail-safe-control, guarantee the safety control of vehicle.

Particularly:

(1) is suitable for when understeer that causes owing to reasons such as road surface and vehicle conditions that occurs that EPS itself is difficult to compensate and ovdersteering problem, guaranteeing that the safety of handling maneuver is finished, the generation that avoids traffic accident.

(2) under the situation that EPS lost efficacy, still can accurately guarantee the stability of running car, avoided the generation of traffic accident.This design has quick, reliable, the low cost and other advantages of operation.

Description of drawings

Fig. 1 is driver-auto model.

Fig. 2 is the main mode of motion of vehicle axis system and automobile.

Fig. 3 is the structured flowchart of state observer.

Fig. 4 is to horizontal surface with vehicle body coordinate and road surface coordinate projection.

Fig. 5 is the structured flowchart of controller of skidding.

Fig. 6 is a wheel stability controller model.

Fig. 7 is the structural representation of electric power-assisted steering apparatus.

Fig. 8 is a control flow chart of the present utility model.

Fig. 9 is a control flow chart of the present utility model.

Figure 10 is a vehicle's center of gravity track emulation curve.

Figure 11 is yaw velocity and the steering curve that does not have behind the EPS malfunction and failure of the utility model device.

Figure 12 is yaw velocity and the turning track curve that has behind the EPS malfunction and failure of the utility model device.

The specific embodiment

Below in conjunction with drawings and Examples the utility model is further specified.

As shown in the figure, the 1st, driver, the 2nd, vehicle, the 3rd, electric direction varying device, the 4th, required side velocity, the 5th, booster torquemoment, the 6th, automobile yaw velocity, the 7th, deflection angle, the 8th, driver's moment of torsion, the 9th, steering dish, the 10th, torque sensor, the 11st, assisted electric machine, the 12nd, four-wheel drives speed-slackening signal, and the 13rd, sensor signal, the 15th, acceleration/accel, the 16th, control unit ECU, 17 torque sensor signals, the 18th, power-assisted electric current.

EPS is made up of torque sensor 10, car speed sensor, current sensor, control unit ECU16, assisted electric machine 11 and speed reduction gearing etc.Wherein, torque sensor 10 and rotary angle transmitter are the sensors of core the most among the EPS.Early stage EPS, particularly low speed type (the power-assisted effect only is provided below a certain speed of a motor vehicle) EPS also has magnetic clutch.When the speed of a motor vehicle surpassed a certain setting value (as 30km/h), cornering resistance moment reduced during owing to high speed, and pilot control bearing circle 9 can turn to, and ECU 16 control magnetic clutchs separately are equivalent to manual steering this moment.The EPS that has power-transfer clutch by ECU 16 control clutchs separately, disconnects the power-assisted effect of electrical motor when breaking down, system enters the manual steering pattern.

EPS working process principle is: torque sensor 10 links together with steering shaft (pinion shaft), during chaufeur steering wheel rotation 9, torque sensor 10 is started working, the displacement that relatively rotates that input shaft and output shaft are produced under the torsion bar effect becomes electric signal and passes to ECU16, ECU16 is according to car speed sensor signal 13, torque sensor signal 17 and x, the hand of rotation and power-assisted electric current 18 sizes of y acceleration/accel 15 decision electrical motors, and the output four-wheel drives speed-slackening signal 12, by distribution to four-wheel braking moment or drive torque, auxiliary EPS turn to or the car body of calming dynamic, thereby control servo-steering in real time.It can be easy to realize that electrical motor provides different power-assisted effects when the different speed of a motor vehicle, and light and flexible when guaranteeing the automobile low speed driving is reliable and stable when running at high speed.Therefore the setting of EPS cornering properties has higher degree of freedom.

Fig. 1 has described six-freedom degree and the system of axes of vehicle body in space motion.With vehicle coordinate is benchmark, and the motion of automobile can be decomposed into: (1) is along the longitudinal movement of x axle; (2) along the sideway movement of y axle; (3) along the perpendicular movement of z axle; (4) around the inclination campaign of x axle; (5) around the luffing of y axle; (6) around the weaving of z axle.The yaw velocity γ 6 and the side slip angle β that it is generally acknowledged automobile are the important parameters of describing motion state of automobile, and these two parameters can characterize the stability of automobile to a great extent.Therefore main consideration and these two the closely-related longitudinal movements of parameter, weaving and sideway movements in the analysis of automobile being carried out road-holding property.

Fig. 2 is the three degree of freedom car model that vehicle axis system and road surface system of axes with Fig. 1 project to horizontal surface, ignores the minute differences of wheel base, and P is the center of gravity of vehicle, l _fBe the distance of P to front axle, l _rBe the distance of P to rear axle, α _fBe front wheel side drift angle, α _rBe rear wheel-side drift angle, δ _fBe the steering angle of wheel flutter, v is the speed of a motor vehicle.F _{X_fl}, F _{X_fr}, F _{X_rr}, F _{X_rl}Be respectively left front, right front, left back, right rear fire longitudinal force F _{Y_fl}, F _{Y_fr}, F _{Y_rl}, F _{Y_rr}Be respectively left front, right front, left back, right rear fire transverse force, I is the rotor inertia of automobile.γ is an automobile yaw velocity 6.β is the car body angle of side slip.N is the vehicle body steering torque that the stressed difference of left and right sides causes.

List following automobile sport equation according to Newton mechanics law:

Longitudinal movement: ma _x=F _{X_fl}+ F _{X_fr}+ F _{X_rl}+ F _{X_rr}

Cross motion: ma _y=F _{Y_fl}+ F _{Y_fr}+ F _{Y_rl}+ F _{Y_rr}

Pendular motion: $I\stackrel{.}{\mathrm{\γ}}=l_f(F_{y\_\mathrm{fl}}+F_{y\_\mathrm{fr}})-l_r(F_{y\_\mathrm{rl}}+F_{y\_\mathrm{rr}})+N$

$N=\frac{d}{2}(-F_{x\_\mathrm{fl}}+F_{x\_\mathrm{fr}}-F_{x\_\mathrm{rl}}+F_{x\_\mathrm{rr}})$

The suffered transverse force of tire can be expressed from the next:

F _{y_fl}＝α _fC _fl F _{y_fr}＝α _fC _fr

F _{y_rl}＝α _rC _rl F _{y_rr}＝α _rC _rr

C _Fl～C _RrIt is respectively each cornering stiffness of taking turns.

Vehicle body speed is decomposed in the direction that wheel center is parallel to the vehicle body system of axes, can try to achieve the speed component of wheel center on the vehicle body coordinate

$v_{x\_\mathrm{fl}}=v\cos \mathrm{\β}-\frac{\mathrm{\γd}}{2}$ v _{y_fl}＝vsinβ+γl _f

$v_{x\_\mathrm{fr}}=v\cos \mathrm{\β}+\frac{\mathrm{\γd}}{2}$ v _{y_fr}＝vsinβ+γl _f

$v_{x\_\mathrm{fl}}=v\cos \mathrm{\β}-\frac{\mathrm{\γd}}{2}$ v _{y_rl}＝vsinβ-γl _r

$v_{x\_\mathrm{rr}}=v\cos \mathrm{\β}+\frac{\mathrm{\γd}}{2}$ v _{y_rr}＝vsinβ-γl _r

Therefore the expression formula of the sideslip angle (oblique speed of advancing of car itself and tire have certain angle) of each tire is:

${\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="1"\; label="f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">f</figure-callout>}1}=\arctan \left(\frac{v_{y\_\mathrm{fl}}}{v_{x\_\mathrm{fl}}}\right)-{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="1"\; label="f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">f</figure-callout>}1}=\arctan \left(\frac{v\sin \mathrm{\&beta;}+\mathrm{\&gamma;}{l}_f}{v\cos \mathrm{\&beta;}-\frac{\mathrm{\&gamma;d}}{2}}\right)-{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="1"\; label="f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">f</figure-callout>}1}$

${\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">r</figure-callout>}1}=\arctan \left(\frac{v_{y\_\mathrm{rl}}}{v_{x\_\mathrm{rl}}}\right)=\arctan \left(\frac{v\sin \mathrm{\&beta;}-\mathrm{\&gamma;}{l}_r}{v\cos \mathrm{\&beta;}-\frac{\mathrm{\&gamma;d}}{2}}\right)$

${\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="2"\; label="f"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">f</figure-callout>}2}=\arctan \left(\frac{v_{y\_\mathrm{fr}}}{v_{x\_\mathrm{fr}}}\right)-{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="f"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">f</figure-callout>}2}=\arctan \left(\frac{v\sin \mathrm{\&beta;}+\mathrm{\&gamma;}{l}_f}{v\cos \mathrm{\&beta;}+\frac{\mathrm{\&gamma;d}}{2}}\right)-{\mathrm{\&delta;}}_{\mathrm{<figure-callout\; id="2"\; label="f"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">f</figure-callout>}2}$

${\mathrm{\&alpha;}}_{\mathrm{<figure-callout\; id="2"\; label="r"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">r</figure-callout>}2}=\arctan \left(\frac{v_{y\_\mathrm{rr}}}{v_{x\_\mathrm{rr}}}\right)=\arctan \left(\frac{v\sin \mathrm{\&beta;}-\mathrm{\&gamma;}{l}_r}{v\cos \mathrm{\&beta;}+\frac{\mathrm{\&gamma;d}}{2}}\right)$

Acceleration/accel under the rest frame X-O-Y is projected to respectively on the x axle, y axle of vehicle body system of axes x-y:

$a_x=-v(\stackrel{.}{\mathrm{\&beta;}}+\mathrm{\&gamma;})\sin \mathrm{\&beta;}+\stackrel{.}{v}\cos \mathrm{\&beta;}$

$a_y=v(\stackrel{.}{\mathrm{\&beta;}}+\mathrm{\&gamma;})\cos \mathrm{\&beta;}+\stackrel{.}{v}\sin \mathrm{\&beta;}$

The acceleration/accel that obtains automobile like this is:

$\stackrel{.}{v}=a=a_x\cos \mathrm{\&beta;}+a_y\sin \mathrm{\&beta;}+v(\stackrel{.}{\mathrm{\&beta;}}+\mathrm{\&gamma;})\sin \mathrm{\&beta;}\cos \mathrm{\&beta;}-v(\stackrel{.}{\mathrm{\&beta;}}+\mathrm{\&gamma;})\cos \mathrm{\&beta;}$

The expression formula of the center of gravity angle of side slip β of electronlmobil is:

$\mathrm{\&beta;}=\arctan \left(\frac{v_y}{v_x}\right)$

As quantity of state, the equation of state of system is with β, γ:

$\stackrel{.}{x}=\mathrm{Ax}+\mathrm{Bu}$

Wherein,

$A=\left[\begin{array}{cc}\frac{({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}}+{\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{\mathrm{mv}}& \frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})-1_r({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{m{v}^2}-1\\ \frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})-1_r({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{I}& \frac{1_f^2({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})+1_r^2({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{\mathrm{Iv}}\end{array}\right]$

$B=\left[\begin{array}{cc}-\frac{{\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}}}{\mathrm{mv}}& 0\\ -\frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})}{I}& \frac{1}{I}\end{array}\right],x=\left[\begin{array}{c}\mathrm{\&beta;}\\ \mathrm{\&gamma;}\end{array}\right],u=\left[\begin{array}{c}{\mathrm{\&delta;}}_f\\ N\end{array}\right]$

Fig. 3 is the structural representation of the β horn shape attitude observer of structure.Be fully the reflection vehicle in the running state of linear zone and inelastic region, whole state variable (vehicle body yaw velocity γ and car body angle of side slip β) all is used for the state estimation of state observer, and with Vehicular yaw cireular frequency γ and car body lateral acceleration a _yOutput feedback variable as state observer.For this reason, in state Observer Design, at first re-construct a _y:

a _y＝v(a ₁₁β+a ₁₂γ+b ₁₁δ+γ)

The equation of state of this state observer and output equation are:

$\stackrel{.}{\hat{x}}=A\hat{x}+\mathrm{Bu}-G(\hat{y}-y)$

$\hat{y}=C\hat{x}+\mathrm{Du}$

In the formula:

C＝[va _11v(a ₁₂+1)]，D＝[vb ₁₁0]，y＝[a _y]

G is the feedback gain matrix of state observer, It is the estimated valve of x.

The ornamental of system is determined by following formula:

$\mathrm{rank}\left[\begin{array}{c}C\\ \mathrm{CA}\end{array}\right]=\mathrm{rank}\left[\begin{array}{cc}{\mathrm{va}}_{11}& v(a_{12}+1)\\ {\mathrm{<figure-callout\; id="11"\; label="va"\; filenames="Y200820075121E00152.png"\; state="\{\{state\}\}">va</figure-callout>}}_{11}^2+v(a_{12}+1)a_{21}& {\mathrm{va}}_{11}{a}_{12}+v(a_{12}+1)a_{22}\end{array}\right]\&NotEqual;0$

Abbreviation gets:

$a_{11}{a}_{22}{a}_{12}+a_{11}{a}_{22}-a_{11}^2-a_{12}^2{a}_{21}-2a_{12}{a}_{21}-a_{21}\&NotEqual;0$

The key of state observer robustness and stability at first is rationally to construct the value of gain matrix G, i.e. the design of G value will consider to reduce the influence of model error, and simultaneously, the eigenwert of matrix A-GC all should be in stable zone.

The matrix of coefficient of state observer is: A+GC

$A+\mathrm{GC}=\left[\begin{array}{cc}{a}_{11}& a_{12}\\ a_{21}& a_{22}\end{array}\right]-\left[\begin{array}{c}{g}_1\\ {\mathrm{<figure-callout\; id="2"\; label="g"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">g</figure-callout>}}_2\end{array}\right]\left[\begin{array}{cc}{\mathrm{va}}_{11}& v(a_{12}+1)\end{array}\right]$

$=\left[\begin{array}{cc}{a}_{11}-{\mathrm{<figure-callout\; id="1"\; label="g"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">g</figure-callout>}}_1{\mathrm{va}}_{11}& a_{12}-g_{1}v(a_{12}+1)\\ a_{21}-g_{2}v{a}_{11}& a_{22}-g_{2}v(a_{12}+1)\end{array}\right]$

The state observer characteristic equation is:

$f\left(s\right)=|\mathrm{SI}-(A-\mathrm{GC})|=\left|\begin{array}{cc}s-a_{11}+{\mathrm{<figure-callout\; id="1"\; label="g"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">g</figure-callout>}}_1{\mathrm{va}}_{11}& -a_{12}+g_{1}v(a_{12}+1)\\ -a_{21}+g_{2}v{a}_{11}& s-a_{22}+g_{2}v(a_{12}+1)\end{array}\right|$

$={\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">s</figure-callout>}}^2+[g_{2}v(a_{12}+1)-a_{22}-a_{11}+{\mathrm{<figure-callout\; id="1"\; label="g"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">g</figure-callout>}}_1{\mathrm{va}}_{11}]s+a_{11}{a}_{22}-a_{12}{a}_{21}$

$+g_2[a_{12}{\mathrm{va}}_{11}-a_{11}v(a_{12}+1)]+g_1[a_{12}v(a_{12}+1)-a_{11}{\mathrm{va}}_{12}]$

$=0$

Can the limit of supposing the system in the stabilized zone be s=-p

The character pair equation is:

f ^*(s)＝(s+p) ²＝s ²+2ps+p ²＝0

F (s) and f ^*(s) corresponding undetermined coefficient gets following equation:

$\left\{\begin{array}{c}{g}_{1}v(a_{12}+1)-a_{22}-a_{11}+{\mathrm{<figure-callout\; id="1"\; label="g"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">g</figure-callout>}}_1{\mathrm{<figure-callout\; id="11"\; label="va"\; filenames="Y200820075121E00152.png"\; state="\{\{state\}\}">va</figure-callout>}}_{11}=2p\\ a_{11}{a}_{12}-a_{12}{a}_{21}-g_2{a}_{11}v+g_1\left[a_{12}v(a_{12}-a_{11}+1)\right]={\mathrm{<figure-callout\; id="2"\; label="p"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">p</figure-callout>}}^2\end{array}\right.$

The group of solving an equation can obtain the feedback gain matrix G of state observer

${\mathrm{<figure-callout\; id="1"\; label="g"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">g</figure-callout>}}_1=\frac{(a_{12}+1){\mathrm{<figure-callout\; id="2"\; label="p"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">p</figure-callout>}}^2+{2a}_{11}p+a_{11}^2+a_{11}{a}_{22}+a_{12}{a}_{21}{+a}_{12}^2{a}_{21}-a_{11}{a}_{22}-a_{11}{a}_{12}{a}_{22}}{(a_{12}^3+2a_{12}^2-a_{11}{a}_{12}^2+a_{11}^2-a_{11}{a}_{12}+a_{12})v}$

${\mathrm{<figure-callout\; id="2"\; label="g"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">g</figure-callout>}}_2=\frac{a_{11}{p}^2-2(a_{12}^2-a_{11}{a}_{12}+a_{12})p-a_{11}^2{a}_{22}+a_{11}{a}_{12}{a}_{21}-a_{12}^2{a}_{22}-a_{12}{a}_{22}+a_{11}{a}_{12}{a}_{22}-a_{11}{a}_{12}^2-a_{11}{a}_{12}+a_{11}^2{a}_{12}}{(a_{11}{a}_{12}^2+a_{11}{a}_{12}-a_{11}^2-a_{12}^4-2a_{12}^3-a_{12}^2)v}$

$G=\left[\begin{array}{c}{g}_1\\ {\mathrm{<figure-callout\; id="2"\; label="g"\; filenames="Y200820075121E00151.png"\; state="\{\{state\}\}">g</figure-callout>}}_2\end{array}\right]$

Feedback gain matrix G changes according to the limit in the stable region of ssystem transfer function, reaches the stable requirement of system for field.

The utility model The controller mainly comprises following components as shown in Figure 4: yaw velocity command generator, torque distribution controller (TDC), tyre slip controller (WSC), and state observer.When EPS lost efficacy, by corner that detects bearing circle 9 or the steering angle δ that torque obtains wheel flutter _fThe yaw velocity γ value that obtains expecting by the yaw velocity command generator, adopt feedforward and PI (proportional integral (PI)) regulating control to regulate and make speed of response accelerate, move accurately, the torque distribution controller distribute driving torque or braking torque by the tyre slip controller action in vehicle body, detect γ by state observer, the β value is formed closed loop control.

When battery-driven car is turned, certain angle is arranged between the direction of tire and actual moving direction:

${\mathrm{\&alpha;}}_f=\mathrm{\&beta;}+\frac{\mathrm{\&gamma;}}{V}{1}_f-{\mathrm{\&delta;}}_f$

The method that improves the operating characteristic of battery-driven car is that the front-wheel of battery-driven car is advanced along the direction at navigation angle, promptly reduces α as far as possible _fTherefore according to following relational expression:

γ ^*＝(δ _f-β)v/l _f

Can obtain best yaw velocity command value γ ^*

Stablize needed moment of rotation size according to car body, front and back wheel group mean allocation drive torque, left and right wheels group complementary allocated steering torque.Concrete torque distribution controller can be realized with the following method:

N＝N _f+N _r

$N_f=\frac{d}{2}(F_{x\_\mathrm{fr}}-F_{x\_\mathrm{fl}})=\frac{1}{2}N$

$N_r=\frac{d}{2}(F_{x\_\mathrm{rr}}-F_{x\_\mathrm{rl}})=\frac{1}{2}N$

$F_{x\_\mathrm{fr}}=F_{x\_\mathrm{fr}}^{*}+\frac{N}{2d}$

$F_{x\_\mathrm{fl}}=F_{x\_\mathrm{fl}}^{*}-\frac{N}{2d}$

$F_{x\_\mathrm{rr}}=F_{x\_\mathrm{rr}}^{*}+\frac{N}{2d}$

$F_{x\_\mathrm{rl}}=F_{x\_\mathrm{rl}}^{*}-\frac{N}{2d}$

N wherein _fBe the front vehicle wheel drive torque, N _rBe the rear wheel drive torque, d is the distance between two-wheeled.

Perhaps adopt the method for only on front-wheel or trailing wheel, carrying out the torque mean allocation.Promptly

$N_f=\frac{d}{2}(F_{x\_\mathrm{fr}}-F_{x\_\mathrm{fl}})=N$ Or $N_r=\frac{d}{2}(F_{x\_\mathrm{rr}}-F_{x\_\mathrm{rl}})=N$

Propulsive effort or braking force are delivered on the wheel via the wheel stability controller.The principle of work of wheel stability controller is, makes up a load observer, allows wheel according to the wheel dynamic model, produces speed according to the propulsive effort command value, even if on the road surface of skidding easily, also can be limited in the allowed band skidding.The wheel stability controller as shown in Figure 6, its closed loop transfer function, can be described as:

$G_{\mathrm{CL}}=\frac{({\mathrm{\&tau;}}_{1}s+1)({\mathrm{\&tau;}}_{2}s+1)}{({\mathrm{\&tau;}}_{1}s+1)({\mathrm{\&tau;}}_{2}s+1)+K(J_n/J-1)}$

τ wherein ₁, τ ₂Be the time constant of low-pass filter, J _nBe folded to the rated value of the whole rotor inertias on the wheel, J is the actual observed value of the whole rotor inertias on the wheel

Work as J=J _nThe time, closed loop transfer function, equals 1; Skidding, the equivalent moment of inertia of car body on wheel will reduce when taking place, and transfer function will so just reduce the driving torque of wheel less than 1, thus further developing of skidding of inhibition.

(list of references Lianbing Li, Shinya Kodama, Yoichi Hori.Anti-Skid Control forEV Using Dynamic Model Error based on Back-EMF Observer.Proc.IECON 2004, Busan, Korea).

By judging the causal relationship of torque sensor and transverse axis acceleration change, can judge whether EPS hinders for some reason and lost efficacy.

Under the situation that the EPS et out of order lost efficacy,, infer the steering angle command value of chaufeur with following formula by detecting the output valve of torque sensor:

${\widetilde{\mathrm{\&delta;}}}_f=K\&CenterDot;T_S$

Wherein Be the guess value of steering angle command value, T _SBe the output valve 17 of torque sensor 10 among Fig. 7, K is the proportionality coefficient that output valve 17 is converted to the guess value of steering angle command value, and its size is greater than 0, near 0.035.

Fig. 8 and Fig. 9 are control flow charts of the present utility model.Among Fig. 8, under the situation of EPS normal operation,, obtain a suitable assisted diversion torque command value, distribute link to be dispensed to each wheel by moment then by detecting torque sensor signal or from the ECU of EPS, reading.Do not turn to attitude to carry out closed loop feedback to vehicle body.

Two kinds of situations of the EPS malfunction and failure situation of Fig. 8 and Fig. 9 all adopt vehicle body to turn to attitude to carry out closed loop feedback.

This device adopts data shown in the table 1, adopts Japanese Nissan series, March number (MARCH) vehicle, according to above-mentioned four wheeler model, $\stackrel{.}{x}=\mathrm{Ax}+\mathrm{Bu}$

Wherein,

$A=\left[\begin{array}{cc}\frac{({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}}+{\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{\mathrm{mv}}& \frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})-1_r({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{m{v}^2}-1\\ \frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})-1_r({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{I}& \frac{1_f^2({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})+1_r^2({\mathrm{<figure-callout\; id="1"\; label="C\; r"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_r+C_{\mathrm{rr}})}{\mathrm{Iv}}\end{array}\right]$

$B=\left[\begin{array}{cc}-\frac{{\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}}}{\mathrm{mv}}& 0\\ -\frac{1_f({\mathrm{<figure-callout\; id="1"\; label="C\; f"\; filenames="Y200820075121E00133.png,Y200820075121E00141.png"\; state="\{\{state\}\}">C</figure-callout>}}_f+C_{\mathrm{fr}})}{I}& \frac{1}{I}\end{array}\right],$

$x=\left[\begin{array}{c}\mathrm{\&beta;}\\ \mathrm{\&gamma;}\end{array}\right],$

$u=\left[\begin{array}{c}{\mathrm{\&delta;}}_f\\ N\end{array}\right]$

Be to carry out the Computer Simulation test on 1.0 the dry asphalt surface at friction coefficient, simulation result is shown in Figure 10～12.

Table 1 vehicle simulation parameters

System parameter	Numerical value	Unit
			Tire
Radius r	0.26	Rice
			Rotor inertia J r	2.5	kgm 2
Resistance wheel F r	11*9.8	Newton
			Reduction ratio Ngear	13.4
Motor torque coefficient C T	0.2122
			Wheel reduced mass M w	Jr/r/r
Cornering stiffness C fl，C fr，C rl，C rr *	2.5×10 4
			Vehicle body
Mass M	1100	kg
			Total length 1 all	3.7	Rice
Rotor inertia I		kgm 2

Center of gravity is to front axle distance 1 f	1.0	Rice
			Center of gravity is to rear axle distance 1 r	1.36	Rice
Wheel base
	1	Lf+lr	Rice
Front wheel spindle is apart from d f **				1.36	Rice
	Hind axle is apart from d r **	1.33	Rice
Air resistance F a				0.552	N s 2/m 2
	Other initial parameters
Vehicle body rate of onset V_init				15	Meter per second
	Wheel rate of onset Vw_init	15	Meter per second
Initial angle of side slip β _ init				0	rad
	Initial yaw velocity γ _ init	0	rad/sec

^*C _Fl, C _Fr, C _Rl, C _RrGet its aviation value.

^*Wheel base is approximate consistent, therefore averages, and all represents with d.

Figure 10 is a vehicle's center of gravity track emulation curve.I is the vehicle's center of gravity track emulation curve that does not add the utility model device among Figure 10, owing to turn to control accuracy low, response has hysteresis, chaufeur is repeatedly revised track, II is to use the vehicle's center of gravity turning track simulation curve behind this clearly demarcated device, and curve turns to control accuracy to improve after showing this device of increase.

Figure 11 is yaw velocity and the steering curve that does not have behind the EPS malfunction and failure of the utility model device.III figure is vehicle velocity V and wheel speed V _W, IV figure is a yaw angle, and V figure is a yaw velocity, and VI figure is a steering angle.VII figure is the vehicle's center of gravity track.

Figure 12 is yaw velocity and the turning track curve that has behind the EPS malfunction and failure of the utility model device.VIII figure is vehicle velocity V and wheel speed V _WIX figure is a yaw angle, and X figure is a yaw velocity, and XI figure is a steering angle.XII figure is the vehicle's center of gravity track.

Figure 10 can not turn to after showing the EPS malfunction and failure.After Figure 12 shows increase the utility model device, still can realize steering operation behind the EPS malfunction and failure by the control of four-wheel torque distribution.

The utility model is on the basis of electric boosting steering system EPS, by distribution to four-wheel braking moment or drive torque, auxiliary EPS turn to or the car body of calming dynamic, guarantee the safety control of vehicle 2.Make automobile be difficult to compensate under the situation of the understeer that causes owing to road surface and vehicle condition reason and ovdersteering problem at original EPS thrashing or EPS itself, chaufeur is made a response, by corner that detects bearing circle 9 or the steering angle δ that torque obtains wheel flutter _fThe γ value that obtains expecting by the yaw velocity command generator, obtain actual yaw velocity γ and angle of side slip β value by state observer, when yaw velocity is passed through the stable control of handling maneuver, output four-wheel braking and driving distributed intelligence, adopt feedforward and pi regulator to regulate the assisted diversion torque, be added to respectively on each wheel by the driving torque or the braking torque of torque distribution controller different sizes, form steering torque, make steering response speed up, move accurately, realize the fail safe function of EPS.This system not only can guarantee the smooth operation of vehicle 2 under the EPS normal circumstances, even also can guarantee the normal direction of rotation of automobile when EPS lost efficacy.

Claims (5)

1, a kind of electric power-assisted steering apparatus, it comprises:

Torque sensor, car speed sensor, current sensor, control unit ECU, assisted electric machine and speed reduction gearing; Bearing circle is linked together by input shaft and torque sensor, and torque sensor also links together with the assisted electric machine, and torque sensor all is connected with control unit ECU with the assisted electric machine; It is characterized in that also comprising:

Wheel stability controller and four wheel drive or lock torque control signal output unit, the consisting of of wheel stability controller:

The yaw velocity command generator is used to obtain the yaw velocity command value;

State observer is used to obtain actual yaw velocity value and angle of side slip value;

The torque distribution controller is added to different big or small driving torque or braking torques respectively on each wheel, forms steering torque, and steering response is speeded up;

The tyre slip controller by regulating the application force between vehicle tyre and road surface, improves the active safety performance of automobile.

2, electric power-assisted steering apparatus according to claim 1 is characterized in that described yaw velocity command generator is by detecting the steering angle δ of wheel flutter _fAnd the difference of angle of side slip β multiply by speed v again divided by between center of gravity and the wheel flutter apart from l _fObtain yaw velocity command value γ *.

3, electric power-assisted steering apparatus according to claim 1 is characterized in that described state observer comprises two inputs, and promptly u and y are output as

Observer contains n integrator and whole state variables is made an estimate, and G is an observer output feedback battle array, its handle

Inverse feedback extremely

The place makes to configuration observer limit, improves its dynamic property, makes as early as possible

Approach zero and introduce, it is a kind of output feedback.

4, electric power-assisted steering apparatus according to claim 1 is characterized in that described torque distribution controller is to stablize needed moment of rotation size according to car body, front and back wheel group mean allocation drive torque, left and right wheels group complementary allocated steering torque; Perhaps only adopt and on front-wheel or trailing wheel, torque is averaged distribution.

5, electric power-assisted steering apparatus according to claim 1, it is characterized in that described tyre slip controller is based on wheel velocity and calculates acceleration/accel and propulsive effort signal, observation car body equivalent moment of inertia makes up external disturbance signal observer, thereby forms the tyre slip controller.

Images