車輛方向盤於不同路面操控模擬

(1)

國立交通大學

機械工程學系

碩士論文

車輛方向盤於不同路面操控模擬

Simulation on Different Road Surface of Vehicle Steering

Wheel

研究生：林志隆

指導教授：成維華教授

(2)

車輛方向盤於不同路面操控模擬

Simulation on Different Road Surface of Vehicle Steering Wheel

研究生：林志隆 Student：Zhi-Long Lin 指導教授：成維華博士 Advisor：Dr.Wei-Hua Chieng 國立交通大學機械工程學系碩士論文 A thesis

Submitted to Department of Mechanical Engineering College of Engineering

National Chiao Tung University in partial Fulfillment of the Requirements

for the Degree of Master

In

Mechanical Engineering

July 2009

Hsinchu, Taiwan, Republic of China

(3)

i

車輛方向盤於不同路面操控模擬

研究生:林志隆指導教授:成維華教授國立交通大學機械工程學系

摘要

本文內容包括直流馬達、馬達驅動電路和微電腦控制器以及車輛方向盤機構，嘗試利用直流馬達的特性，模擬車輛在不同道路上行駛轉向，車輛方向盤上所反映的力矩變化，利用車輛動力學，我們可以藉由車輛的特徵參數及車輛行駛狀態來模擬，車輛行駛在不同的道路上，車輛方向盤應的回饋力矩變化。更可以透過更改微電腦控制器中車輛的特徵參數和車輛行駛狀態，模擬不同車輛在不同路面上轉向時車輛方向盤力矩變化。未來可發展為線控轉向系統其中的力回饋方向盤系統，以取代傳統轉向系統，透過力回饋馬達將路面狀況回授給駕駛人。關鍵字：直流馬達、車輛方向盤、回饋力矩、線控轉向

(4)

ii

Simulation on Different Road Surface of Vehicle Steering Wheel

Student: Zhi-Long Lin Advisor: Dr. Wei-Hua Chieng

Department of Mechanical Engineering National Chiao Tung University

Abstract

This article content including the DC motor, the DC motor driving circuit and the microcomputer controller as well as the vehicles steering wheel organization, the attempt use DC motor's characteristic, simulates the vehicles to go on the different path changes, on the vehicles steering wheel reflected the moment variation. Using the vehicles dynamics, we may simulates because of vehicles' characteristic parameter and the vehicles drive condition, the vehicles travel on the different path, the vehicles steering wheel should respond how many moment variation.

In this article, we can change microcomputer controller vehicles' characteristic parameter and vehicles drive condition, easier simulation different vehicles go the vehicles steering wheel moment variation which in different road surfaces.

Keywords: DC motor, steering wheel of vehicle, feedback torque, steering-by-wire (SBW)

(5)

iii

誌謝

首先真誠的感謝成維華教授和鄭時龍博士，感謝兩位老師在學生兩年的碩士生涯中的指導，孜孜不倦的教導求學問的態度與研究的方式，讓學生在這兩年收穫良多，再度的感謝兩位老師的付出與指導。感謝實驗室嘉豐學長、秉霖學長和向斌學長，給予的指導，感謝強哥、文彬、洋豪和小賴學長，在學期間對學生的照顧，感謝學磊、星雲和富源學長給予業界上的意見。感謝同學們文祥、冠今、安鎮和之浩，因為有你們讓我在學期間不至於感到太過孤單。這一路上受到幫助的實在太多，需要感謝的人無法一一提名道謝，謝謝這一路有你們的幫忙，我才能夠順利完成學業。這兩年來的日子，點點滴滴不但豐富了我的學識，也豐富了我的視野，最後感謝我的家人願意支持我在服完役之後再度投入學生生活來完成我的夢想，我想沒有家人的支持，沒有今天的我，深深的感謝我的父母。林志隆僅於 2009.08

(6)

iv

List of Tables

TABLE5.1PARAMETER OF STEER-BY-WIRE MODEL ... 68

TABLE5.2VEHICLE PARAMETER ... 68

(10)

viii

List of Figures

FIGURE1.1THE PHYSICAL MODEL OF A TIRE ... 36

FIGURE1.2SCHEMATIC DIAGRAM OF THE SYSTEM ... 36

FIGURE1.3SCHEMATIC BLOCK DIAGRAM OF THE SYSTEM ... 37

FIGURE2.1ILLUSTRATION OF TYPICAL STEERING SYSTEMS ... 37

FIGURE2.2STEER ROTATION GEOMETRY AT THE ROAD WHEEL ... 38

FIGURE2.3FRONT VIEW OF THE CAMBER ANGLE ... 38

FIGURE2.4FRONT VIEW OF THE KINGPIN INCLINE ANGLE ... 39

FIGURE2.5TOE-IN AND TOE-OUT OF A CAR ... 39

FIGURE2.6A POSITIVE AND NEGATIVE CASTER OF A CAR ... 40

FIGURE2.7TOP VIEW OF THE SIDE SLIP ANGLE ... 40

FIGURE2.8SAE TIRE FORCE AND MOMENT AXIS SYSTEM ... 41

FIGURE2.9FORCES AND MOMENTS ACTING ON WHEEL ... 41

FIGURE2.10MOMENT BY VERTICAL FORCE ACTING ON K.P.I. ANGLE ... 42

FIGURE2.11MOMENT BY VERTICAL FORCE ACTING ON CASTER ANGLE ... 42

FIGURE2.12STEERING MOMENT PRODUCED BY TRACTION FORCE ... 43

FIGURE2.13STEERING MOMENT PRODUCED BY LATERAL FORCE ... 43

(11)

ix

FIGURE2.15EQUIVALENT DYNAMIC MODEL OF STEERING SYSTEM ... 44

FIGURE3.1MECHANISM DIAGRAM OF MECHANICAL STEERING SYSTEM ... 45

FIGURE3.2MATH MODEL OF THE MECHANICAL STEERING SYSTEM ... 45

FIGURE3.3STEERING WHEEL SYSTEM AND FRONT WHEEL SYSTEM ... 46

FIGURE3.4MATH MODEL OF STEERING SYSTEM ... 46

FIGURE3.5STEER BY WIRE SYSTEM ... 47

FIGURE3.6CONTROL BLOCK DIAGRAM OF SBW ... 48

FIGURE3.7SAE VEHICLE AXIS SYSTEMS. ... 49

FIGURE3.82DOF OF BICYCLE MODEL ... 49

FIGURE3.9SIMPLIFIED MODEL OF STEERING SYSTEM ... 50

FIGURE3.10REACTION TORQUE ON DIFFERENT ROAD ... 50

FIGURE3.11REACTION TORQUE OF DIFFERENT VEHICLE VELOCITY ... 51

FIGURE3.12LQG CONTROL STRUCTURE ... 51

FIGURE4.1POLE/ZERO MAP OF MECHANICAL MODEL ... 52

FIGURE4.2RED FRAME ON MECHANICAL MODEL ... 52

FIGURE4.3POLE/ZERO MAP OF STEER-BY-WIRE MODELS ... 53

FIGURE4.4RED FRAME ON STEER-BY-WIRE MODELS ... 53

(12)

x

FIGURE4.6RED FRAME ON SBW MODEL WITH LQG CONTROL ... 54

FIGURE4.7SBW MODEL ROOT LOCUS ... 55

FIGURE4.8SBW MODEL WITH LQG ROOT LOCUS ... 55

FIGURE4.9SIMULATION WITH LQG, WHEN INPUT IS TORQUE HUMAN ... 56

FIGURE4.10SIMULATION WITH LQG, WHEN INPUT IS EXTERNAL TORQUE ... 57

FIGURE4.11CASE 1: SIMULATION ON DIFFERENT ROAD ... 58

FIGURE4.12CASE1: VEHICLE LOCUS ON DIFFERENT ROAD ... 58

FIGURE4.13CASE 1: SIMULATION OF DIFFERENT VEHICLE VELOCITY ... 59

FIGURE4.14CASE1: VEHICLE LOCUS OF DIFFERENT VEHICLE VELOCITY ... 59

FIGURE4.15CASE 1: SIMULATION WITH AND WITHOUT LQG CONTROL ... 60

FIGURE4.16CASE 1: VEHICLE LOCUS WITH AND WITHOUT LQG CONTROL ... 60

FIGURE4.17CASE 1:SIMULATION WITH AND WITHOUT LQG CONTROL ... 61

FIGURE4.19CASE 2: SIMULATION ON DIFFERENT ROAD ... 62

FIGURE4.20CASE 2: VEHICLE LOCUS ON DIFFERENT ROAD ... 62

FIGURE4.21CASE 2: SIMULATION OF DIFFERENT VEHICLE VELOCITY ... 63

FIGURE4.22CASE 2: VEHICLE LOCUS OF DIFFERENT VEHICLE VELOCITY ... 63

(13)

xi

FIGURE4.25CASE 2: SIMULATION WITH AND WITHOUT LQG CONTROL ... 65

FIGURE4.27CASE 3: SIMULATION WITH AND WITHOUT LQG... 66

(14)

1

Chapter 1 Introduction

1.1.Motive and objectives

We want to build the steer-by-wire model to replace the traditional steering model. The steer-by-wire system used the electronic signal to connect the steering wheel and front wheel reduce the steering column on vehicle can made the vehicle cost down, weight light and faster. So the project‟s objective is build the steer by wire (SBW) system control structure. In future we control steering of vehicle from the control structure.

1.2. Literature Review

The feedback torque of steering wheel is from the steering system of vehicles when the vehicle corner. So we research the steering system of vehicles to build the model of vehicles. From the steering system we know the feedback torque of steering wheel produced from the force on the tire of front wheels.

From 1931 year, Bradley and Allen being to research the dynamic of vehicle tires. Now, we have several ways to build the model of tire.

1.2.1. Faila’s model

Faila used the basis of Fromm‟s model obtain the equation of lateral force, aligning torque and slip angle.[1]

2 2 3 3 4

2 3

1 1 1

2 2 2

1 1

tan tan tan

2 8 _m 96 _m K l K l K l F p b p b





   (1- 1) 3 2 4 3 5 4 6 2 3 4 1 1 1 1 2 2 2 3 3 3 1 1 1

tan tan tan tan

12 16 _m 64 _m 768 _m K l K l K l K l M p b p b p b





    (1- 2)

(15)

2 0 1 3 3 0 1 12 K K l K k



  (1- 3)





0 2 1 E b K d



  (1- 4) Where: F = lateral force. M = Feedback torque.

E =Young's Modulus of tire surface stuff.



= Poisson ratio.

b =Width of contact ground surface.

d = Width of tire surface rubber.

k = Spring constant.

pm = Max pressure of contact ground.

l = Length of contact ground surface.



= friction coefficient between ground and tire surface.

β

= Slip angle of tire.

1.2.2. Magic Formula

Bakker and Pacejka obtain “Magic Formula” on the basis of experiment. Express the relationship of lateral force, traction force and feedback torque.[2][3][4]

   

v

Y X  y x S (1- 5)

 

sin



arctan



arctan

 





(16)

3

h

x X S _{(1- 7)}

Where:

Y(X) = Generalized force.

X = Generalized offset.

B = Stiffness factor.

C = Shape factor.

D = Peak factor.

E = Curvature factor.

Sv,Sh = Initial offset of slip angle.

1.2.3. Mitschke’s model

Mitschke‟s idea is the feedback torque includes two parts. One is the lateral force and tire trail to produce, which is in direct ratio with centrifugal acceleration. Another produced by the kingpin incline angle with offset, which unrelated with velocity of vehicle.[5]





Z y M P    _{(1- 8)} 2 y v b P m R l  (1- 9) 2 2 1 1 2 f k b k a l m v k k l R



 _        (1- 10) Where:

Mz = Moment produced by tire trail.

(17)

4



 _{= Pneumatic trail.}



 _{= Mechanical trail.}

m = Mass of vehicle.

b = Longitudinal distance from center of gravity to rear axle.

l = Wheelbase.

R = Radius of turn.

v = Vehicle velocity.

a = Longitudinal distance from front axle to center of gravity.

k1 = Cornering stiffness of front wheel.

k2 = Cornering stiffness of rear wheel.



f = Steer angle of front wheel.

The aligning torque produced by kingpin incline angle, which unrelated with velocity of vehicle. At the high speed has the same torque at low speed. This is the main source of feedback torque at low speed.

sin 2

sin

2

A f

Q D

M











(1- 11) Where:

MA = Moment produced by kingpin incline angle.

Q = The offset of kingpin.

D = Load of tire.



 _{= Kingpin incline angle.}

(18)

5 Z A h f M M M i   _{(1- 12)} Where: if = Gear ratio.

1.2.4. UniTire model

Guo and his research team obtain “UniTire model” on the basis of experiment. The physical model of a tire is simplified as shown in Figure1.1.[6][7]













2 2 3 1 1 2 0 1 2 1 1 exp 12 exp x x e e x x x z y y y z z y x c x c F E E D D D D D D F F F F F F M F D X F Y



 







      _  _  _ _               (1- 13) Where: F = Total force. Fx = Traction force. Fy = lateral force. E1 = Curvature factor. Dx0, De, D1, D2 = Structure parameter of Dx.



x = Longitudinal slip ratio.



y = Lateral slip ratio.

(19)

6

Xc = The longitudinal out of shape produced by Fx.

Yc = The lateral out of shape produced by Fy.



x = The friction coefficient on longitudinal surface.



y = The friction coefficient on lateral surface.

1.2.5. The model of vehicles

From fundamentals of vehicles dynamics, we can analyze the feedback torque of steering wheel. We will detail illustration on next chapter.

1.3. Structure of The Steering Wheel

A feedback dynamic gear force is applied physically to the steering wheel hardware by a motor actuator. A command signal for a motor-force actuator is generated from a virtual vehicle handling dynamics model that is simulated in the real time by a very fast digital signal processor. The inputs to the real-time vehicle dynamic simulation model are front wheel steering angles driven through a steering system by a driver. Figure1.2 and Figure1.3 illustrate some of these.

(20)

7

Chapter 2 Vehicle Steering System

2.1. The Steering System

2.1.1. Introduction of the Steering System

The design of the steering system has an influence on the direction response behavior of a motor vehicle that is often not fully appreciated. The function of the steering system is to steer the front wheels response to driver command inputs in order to provide overall directional control of the vehicle. However, the actual steer angle achieved are modified by the geometry of the suspension system, the geometry and reactions within the steering system, and in the case of front-wheel-drive (FWD), the geometry and reactions from the drive train. These phenomena will be examined in this section first as a general analysis of a steering system and then by considering the influence of front- wheel-drive.

2.1.2. The steering linkages

The steering system used on motor vehicles vary widely in design [9][10][11], but are functionally quite similar. Figure2.1 illustrates some of these.

The steering wheel connects by shaft, universal joints, and vibration isolators to the steering gearbox whose purpose is to transform the rotary motion of the steering wheel to a translational motion appropriate for steering the wheels. The rack-and-pinion system consists of a linearly moving rack and pinion, mounted on the firewall or a forward cross member, which steers the left

(21)

8

and right wheels directly by a tie-rod connection. The tie-rod linkage connects to steering arms on the wheels, thereby controlling the steer angle. With the tie-rod located ahead of the wheel center, as shown in Figure2.1, it is a forward-steer configuration.

The steering gear box is an alternative design used on passenger cars and light trucks. It differs from the rack-and-pinion in that a frame-mounted steering gearbox rotates a pitman arm which controls the steer angle of the left and right wheels through a series of relay linkages and tie rods, the specific configuration of which varies from vehicle to vehicle. A rear-steer configuration is show in the figure, identified by the fact that the tie-rod linkage connects to the steering arm behind the wheel center.

Between these two, the rack-and-pinion system has been growing in popularity for passenger cars because of the obvious advantage of reduced complexity, easier accommodation of front-wheel-drive systems, and adaptability to vehicle without frames. The primary functional difference in the steering systems used on heavy truck is the fact that the frame-mounted steering gearbox steers the left road wheel through a longitudinal drag line, and the right wheel is steered from the left wheel via a tie-rod linkage [9].

The gearbox is the primary means for numerical reduction between the rotational input from the steering wheel and the rotational output about the steer axis. The steering wheel to road wheel angle ratios normally vary with angle, but have nominal values on the order of 15 to 1 in passenger cars, and up to as much as 36 to 1 with some heavy trucks. Initially all rack-and-pinion gearboxes had a fixed gear ratio, in which case any variation in ratio with steer angle was

(22)

9

achieved through the geometry of the linkages. Today, rack-and-pinion systems are available that vary their gear ratio directly with steer angle.

2.2. Tire Angle Geometry

2.2.1. Front wheel geometry

The important elements of a steering system consist not only of the visible linkages just described, but also the geometry associated with the steer rotation axis at the road wheel. The geometry determines the force and moment reaction in the steering system, affecting its overall performance. The important features of the geometry are shown in Figure2.2.

The steer angle is achieved by rotation of the wheel about a steer rotation axis. Historically, this axis has the name “kingpin” axis, although it may be established by ball joint or the upper mounting bearing on a strut. The axis is normally not vertical, but may be tipped outward at the bottom, producing a lateral inclination angle (kingpin inclination angle; K.P.I.) in the range of 10-15degrees on passenger cars.

It is common for the wheel to be offset laterally from the point where the steer rotation axis intersects the ground. The lateral distance from the ground intercept to the wheel centerline is the offset at the ground and is considered positive when the wheel is outboard of the ground intercept. Offset may be necessary to obtain packaging space for brakes, suspension, and steering components. At the same time, it adds “feel of the road” and reduces static steering efforts by allowing the tire to roll around an arc when it is turned [12]. Caster angle result when the steer rotation axis is inclined in the

(23)

10

longitudinal plane. Positive caster places the ground intercept of the steer axis ahead of the center of tire contact. A similar effect is created by including a longitudinal offset between the steer axis and the spin axis of the wheel, although this is only infrequently used. Caster angle normally ranges from 0 to 5 degrees and may vary with suspension deflection.

2.2.2. Camber angle

Camber angle is the angle made by the wheel of an automobile; specifically, it is the angle between the vertical axis of the wheel and the vertical axis of the vehicle when viewed from the front or rear. It is used in the design of steering and suspension. If the top of the wheel is farther out than the bottom (that is, away from the axle), it is called positive camber; if the bottom of the wheel is farther out than the top, it is called negative camber. The camber angle can be recognized better in a front view, as shown in Figure2.3.

The different camber angle change the touch point and the pivot point of tire with ground, direct effect the traction and abrasion of tire. Then change the achieve force distribute on axle from vehicle weight, avoid axle produces exceptionally abrasion.

Besides, the camber can be cancel the angels change from vehicle weight on suspension system to become deformed and the gap of active surfaces. Camber also effects the car drive direction, like the motorcycle incline the body to corner. So the camber of the right and left wheels must equivalent can‟t affect the cars straight-line after the balance of force. And combine toe angle, greater straight-line stability and avoid tire exceptionally abrasion. Increase the negative

(24)

11

camber needs combine increase toe-out. Increase the positive camber needs combine increase toe-in.

2.2.3. Kingpin inclination angle(Lateral inclination angle)

The angle is in front elevation between the steering axis and the vertical. The K.P.I. angle can be recognized better in a front view, as shown in Figure2.4. The K.P.I. can make the vehicle weight uniform distribution on the axis; protect the bearing to impair to hard. The angle can make the cornering force uniform and steering easily. Otherwise, if the angle is zero, then the vehicle weight and reacting force from ground will produce more lateral shear force, make the bearing to impair to easy and hard to cornering. Besides, the angle is the source of the aligning torque after the front wheel corning. The angle usually can‟t change from the vehicle suspension design started.

2.2.4. Toe angle

Toe is the symmetric angle that each wheel makes with the longitudinal axis of the vehicle, as a function of static geometry, and kinematic and compliant effects. The toe angle can be recognized better in a top view, as shown in Figure2.5. This can be contrasted with steer, which is the ant symmetric angle, i.e. both wheels point to the left or right, in parallel (roughly). Positive toe, or toe in, is the front of the wheel pointing in towards the centerline of the vehicle. Negative toe, or toe out, is the front of the wheel pointing away from the centerline of the vehicle. Toe can be measured in linear units, at the front of the tire, or as an angular deflection.

(25)

12

wheels are closer together than the backs of the front wheels) provides greater straight-line stability at the cost of some sluggishness of turning response, as well as a little more tire wear as they are now driving a bit sideways. On front wheel drive cars, the situation is more complex.

Toe is always adjustable in production automobiles, even though caster angle and camber angle are often not adjustable. Maintenance of front end alignment, which used to involve all three adjustments, currently involves only setting the toe ; in most cases, even for a car in which caster or camber are adjustable, only the toe will need adjustment.

One related concept is that the proper toe for straight line travel of a vehicle will not be correct while turning, since the inside wheel must travel around a smaller radius than the outside wheel; to compensate for this, the steering linkage typically conforms more or less to Ackermann steering geometry, modified to suit the characteristics of the individual vehicle.

It should be noted that individuals who decide to adjust their car's static ride height, either by raising or lowering, should immediately have the car properly aligned. The common misconception is that Camber angle causes an increased rate of tire wear, when in fact camber's contribution to tire wear is usually only visible over the entire life of the tire.

2.2.5. Caster angle

Caster angle is the angular displacement from the vertical axis of the suspension of a steered wheel in a car, bicycle or other vehicle, measured in the longitudinal direction. It is the angle between the pivot line (in a car - an

(26)

13

imaginary line that runs through the center of the upper ball joint to the center of the lower ball joint) and vertical. The caster angle can be recognized better in a lateral view, as shown in Figure2.6. Car racers sometimes adjust caster angle to optimize their car's handling characteristics in particular driving situations.

The pivot points of the steering are angled such that a line drawn through them intersects the road surface slightly ahead of the contact point of the wheel. The purpose of this is to provide a degree of self-centering for the steering - the wheel casters around so as to trail behind the axis of steering. This makes a car easier to drive and improves its straight line stability (reducing its tendency to wander). Excessive caster angle will make the steering heavier and less responsive, although, in racing, large caster angles are used to improve camber gain in cornering. Caster angles over 10 degrees with radial tires are common. Power steering is usually necessary to overcome the jacking effect from the high caster angle.

The steering axis (the dotted line in the diagram above) does not have to pass through the center of the wheel, so the caster can be set independently of the mechanical trail, which is the distance between where the steering axis hits the ground, in side view, and the point directly below the axle. The interaction between caster angle and trail is complex, but roughly speaking they both aid steering, caster tends to add damping, while trail adds 'feel', and return ability. In the extreme case of the shopping trolley (shopping cart in the US) wheel, the system is undammed but stable, as the wheel oscillates around the 'correct' path. The shopping trolley/cart setup has a great deal of trail, but no caster. Complicating this still further is that the lateral forces at the tire do not act at the

(27)

14

center of the contact patch, but at a distance behind the nominal contact patch. This distance is called the pneumatic trail and varies with speed, load, steer angle, surface, tire type, tire pressure and time. A good starting point for this is 30 mm behind the nominal contact patch

2.2.6. Slip angle

Slip angle is the angle between a rolling wheel's actual direction of travel and the direction towards which it is pointing. This slip angle results in a force perpendicular to the wheel's direction of travel (the cornering force). This cornering force increases approximately linearly for the first few degrees of slip angle, and then increases non-linearly to a maximum before beginning to decrease.

A non-zero slip angle arises because of deformation in the tire carcass and tread. As the tire rotates, the friction between the contact patch and the road result in individual tread 'elements' ( infinitely small sections of tread ) remaining stationary with respect to the road. If a side-slip velocity u is introduced, the contact patch will be deformed. As a tread element enters the contact patch the friction between road and tire means that the tread element remains stationary, yet the tire continues to move laterally. This means that the tread element will be „deflected‟ sideways. In reality it is the tire/wheel that is being deflected away from the stationary tread element, but convention is for the co-ordinate system to be fixed around the wheel mid-plane. The slip angle can be recognized better in a top view, as shown in Figure2.7.

(28)

15

not distributed equally, the slip angles of each tire will be different. The ratios between the slip angles will determine the vehicle's behavior in a given turn. If the ratio of front to rear slip angles is greater than 1:1, the vehicle will tend to understeer, while a ratio of less than 1:1 will produce oversteer. Actual instantaneous slip angles depend on many factors, including the condition of the road surface, but a vehicle's suspension can be designed to promote specific dynamic characteristics. A principal means of adjusting developed slip angles is to alter the relative roll couple (the rate at which weight transfers from the inside to the outside wheel in a turn) front to rear by varying the relative amount of front and rear lateral load transfer. This can be achieved by modifying the height of the Roll centers, or by adjusting roll stiffness, either through suspension changes or the addition of an anti-roll bar.

Because of asymmetries in the side-slip along the length of the contact patch. The resultant force of this side-slip occurs behind the geometric center of the contact patch, a distance described as the pneumatic trail, and so creates a torque on the tire.

2.2.7. Formulation of geometry

Before we analyze the geometry, we define the geometry meaning of each term in fundamental dynamics equation.

Nomenclature

d : Kingpin offset at the ground

Fxr : Traction force on right-front wheel

(29)

16

Fyr : Lateral force on right-front wheel

Fyl : Lateral force on left-front wheel

Fzr : Vertical force on right-front wheel

Fzl : Vertical force on left-front wheel

Mzr : Aligning torque on right-front wheel

Mzl : Aligning torque on left-front wheel

MT : Moment produced by traction force

ML : Moment produced by lateral force

MV : Moment produced by vertical force

MV,KPI : Moment produced by vertical force acting on kingpin inclination

angle

MV,caster : Moment produced by vertical force acting on caster angle

MAT : Moment produced by aligning torque

Ts : The feedback torque of steering system

Th : The torque of driver steering

r : Tire radius

Rw : Steering wheel radius

λ : Kingpin inclination angle

υ : Caster angle

δ : Tire steer angle



: Steering wheel angle



: Slip angle

Is : Moment of inertia of steering shaft

(30)

17

Ks : Elasticity coefficient of steering system

Ih : Moment of inertia of steering wheel

Ch : Damping coefficient of steering wheel

2.3. Steering System Forces and Moments

The forces and moments imposed on the steering system emanate from those generated at the tire-road interface. The SAE has selected a convention by which to describe the force on a tire, as shown in Figure2.8. Assume that the forces are measured at the center of the contact with the ground and provide a convenient basis by which to analyze steering reactions.

The ground reactions on the tire are described by three forces and moments, as follow:

1. Vertical force 2. Traction force 3. Lateral force 4. Aligning torque

5. Rolling resistance moment 6. Overturning moment

The reaction in the steering system is described by the moment produced on the steer axis, which must be resisted to control the wheel steer angle. Ultimately, the sum of moments from the left and right wheels acting through the steering linkages with their associated ratios and efficiencies account for the steering-wheel torque feedback to the driver.

(31)

18

Figure2.9 shows the three forces and moments acting on a right-hand road wheel. Each will be examined separately to illustrate its effect on the steering system.

2.3.1. Vertical force

The vertical load, Fz, acts vertically upward on the wheel and by SAE

convention is considered a positive force. Because the steering axis is inclined, FZ has a component acting to product a moment attempting to steer the wheel.

The moment arises from both the caster and kingpin inclination angles. Assuming small angles and neglecting camber of the wheel as it steers, the total moment from the two can be approximated by:





sin sin





sin cos

V zl zr zl zr

M

 

F



F

d







F



F

d





(2-1)

Where:

MV = Total moment from left and right wheels.

Fzl, Fzr = Vertical load on left and right wheels.

d = Kingpin offset at the ground.



= Kingpin inclination angle.



= Steer angle.



= Caster angle.

The first expression on the right side of the above equation arises from kingpin inclination angle, and the last from caster angle. The source of each of these moments is most easily visualized by considering the effects of kingpin inclination angle and caster angle separately.

(32)

19

The vertical force acting on kingpin inclination angle, illustrated in Figure2.10, results in a sine angle force component, which normally acts lateral on the moment arm when the wheel is steered. The moment is zero at zero steer angle. When steering, both sides of the vehicle lift, an effect which is often described as the source of the centering moment.





sin



sin



,  F F d 

M_V_inclinatio_n _zl _zr (2-2)

The caster angle results in a sin angle force component, which nominally acts forward on the moment arm as shown in Figure2.11. The moment on the left and right wheels are opposite in direction.





sin



cos



,  F F d 

M_V_caster _zl _zr (2-3)

When steer angle, one side of the axle lift and the other drops, so that the net moment produce depends also on the roll stiffness of the front suspension as it influences the left and right wheel roads.

2.3.2. Traction force

The traction force, Fx, acts on the kingpin offset to produce a moment as

shown in Figure2.12. Then net moment is:





T xl xr

M



F



F

d

(2-4)

Where:

(33)

20

The left and right moments are opposite in direction and tend to balance through the relay linkage. Imbalances, such as may occur with a tire blowout, brake malfunction, or split coefficient surfaces, will tend to produce a steering moment which is dependent on the lateral offset dimension.

2.3.3. Lateral force

The lateral force, Fy, acting at the tire center produces a moment through

the longitudinal offset resulting from caster angle, as shown in Figure2.13. The net moment produced is:





tan

L yl yr

M



F



F

r



(2-5)

Where:

Fyl,Fyr = Lateral force on left and right wheels (positive forward).

r = Tire radius.

The lateral force is generally dependent on the steer angle and cornering condition, and with positive caster produces a moment attempting to steer the vehicle out of the turn.

2.3.4. Aligning torque

The aligning torque, M, acts vertically and may be resolved into a component acting parallel to the steering axis. Since moments may be translated without a change in magnitude, the equation for the net moment is:





2 2

cos

AT zl zr

M



M



M







(2-6)

(34)

21

Myl,Myr = Aligning torque on left and right wheels.

Under normal driving conditions, the aligning torques always act to resist any turning motion, thus their effect is under steer. Only under high braking conditions do they act in a contrary fashion.

2.3.5. Rolling resistance moment and Overturning moment

These moments at most only have a sine angle component acting about the steer axis. They are second-order effects and are usually neglected in analysis of steering system torques.

2.4. Torque feedback on the steering wheel

These equations, (2-1),(2-4), (2-5), (2-6), are the moments of steering axis, we can add direct them to get the total moment, TS, on the kingpin axis:

s V L T AT

T



M



M



M



M

(2-7)

Then we can through the steering system model, as shown Figure2.14, and calculate the force on the steering wheel by the way of the total moment. We can the steering system model to simplify, which equivalent that, as show Figure2.15.





s s s s

I







C







K

 





T

(2-8)

Where:

Is = Moment of inertia of steering shaft.

Cs = Damping coefficient of steering shaft.

(35)

22



= Steering wheel angle.

(36)

23

Chapter 3 Math Model of Steering System

3.1 Nomenclature of the math model

Before we analyze the model of steering system, we define the symbols meaning of each term in fundamental equation.

Symbol Description

J1 Steering wheel moment of inertia

L1 Steering wheel motor inductance

R1 Steering wheel motor resistances

KΦ1 Steering wheel motor voltage constant

Kτ1 Steering wheel motor torque constant

J2 Front wheel moment of inertia

B2 Front wheel damping

K2 Front wheel torsion stiffness

L2 Front wheel motor inductance

R2 Front wheel motor resistances

KΦ2 Front wheel motor voltage constant

Kτ2 Front wheel motor torque constant

K Steering column torsion stiffness

B Steering column damping

n Steering angle ratio

Q Steering torque ratio

Fx Longitudinal Force

(37)

24

Mz Moment of z axis

m Vehicle‟s mass

Vx Longitudinal velocity of vehicle

Vy Lateral velocity of vehicle

Vz Normal velocity of vehicle

V _x Longitudinal acceleration of vehicle V _y Lateral acceleration of vehicle

q Pitch velocity of vehicle‟s y axis

γ Yaw velocity of vehicle‟s z axis

Iz vehicle yaw moment of inertia

C_αf Front tire cornering stiffness coefficient C_αr Rear tire cornering stiffness coefficient

α_f Slip angles for front tires

α_r Slip angles for rear tires

C_sf Front tire longitudinal stiffness coefficient C_sr Rear tire longitudinal stiffness coefficient

s Slip ratio

Vs Slip velocity

μ0 Road fiction coefficient

As Adhesion reduction coefficient

z Dugoff coefficient

(38)

25

3.2 Mechanical steering system

We can obtain the math model as Figure3.2 from the mechanism as Figure3.1 of traditional steering system.

J₁dθ12 d2_t + B1 dθ2 dt = Thum − B dθ1 dt − dθ2 dt − K θ1 − θ2 (3- 1) J₂dθ22 d2_t + B2 dθ2 dt + K2θ2 = Text + B dθ1 dt − n dθ2 dt + K θ1 − θ2 (3- 2) And we cut off the steering column made the steering system become two subsystems as shown in Figure3.3. One of the subsystems is the steering wheel system to feedback the force of the tire on the road. Another of the subsystems is the front wheel system to steer the vehicle. We can obtain the equivalent equations. J₁dθ12 d2_t + B1 dθ2 dt = Thum + T1 (3- 3) J2dθ2 2 d2_t + B2 dθ2 dt + K2θ2 = Text + T2 (3- 4)

Than we need the device to feedback the force of tire and to steer the front wheel. The math model of steering wheel and front wheel system is as shown in Figure3.4.

3.3 Steer-by-wire system

The steer-by-wire system is used the motor replace steering column as shown in Figure3.5. So we can obtain the equivalent equation of the math model of mechanical steering system.

J₁dθ12 d2_t + B1 dθ1 dt = Kτ1i1 + Thum (3- 5) L₁di1 dt + R1i1 + K∅1 dθ1 dt = −B dθ1 dt − n dθ2 dt − K θ1 − nθ2 (3- 6)

(39)

26 J₂dθ22 d2_t + B2 dθ2 dt + K2θ2 = Kτ2i2 + Text (3- 7) L₂di2 dt + R2i2 + K∅2 dθ2 dt = QB dθ1 dt − n dθ2 dt + QK θ1 − nθ2 (3- 8)

Control block diagram of SBW as show in Figure3.6. Torque human is the driver steering torque on steering wheel. It is the one of input of SBW system. Another of input of SBW system is the torque external. It is mean the total force on front wheel tire. Then we known the steering wheel angle and front wheel angle by angle senor on steering wheel and front wheel angle. The SBW system included steering wheel, steering wheel motor, front wheel motor and front wheel of steering system on vehicle. The ratio of front wheel angle between steering wheel angles is n. The ratio of torque human between external torques is Q.

3.4 Torque of the external environment

We feel the condition of the road through the tires, steering column and steering wheel from the external environment. Now we try to simulate actual driving conditions caused by the external environment torque. Before we analysis the vehicle dynamic, we defined the vehicle axis system as shown in Figure3.7and 2DOF of bicycle model as shown in Figure3.8.

First we used the dynamical equation of vehicle.

F_x = m V _x + qV_z − γV_y (3- 9)

F_y = m V _y + γV_x (3- 10)

M_z = I_zγ (3- 11)

(40)

27 effect. Then we simplify these equations.

F_x = −mγV_y (3- 12)

At the small slip angle, the lateral force at front and rear tire can be derived as

F_yf = −C_αfα_f (3- 13)

F_yr = −C_αrα_r (3- 14)

From definition of longitudinal stiffness of tire we have

Fxf = Csfsf (3- 15)

F_xr = C_srs_r (3- 16)

We can obtain the dynamical equation.

F_x = F_xf + F_xr = −mγV_y = C_sfs_f + C_srs_r (3- 17)

F_y = F_yf + F_yr = m V _y + γV_x = −C_αfα_f− C_αrα_r (3- 18) M_z = F_yfL_f − F_yrL_r = I_zγ = −C_αfα_fL_f + C_αrα_rL_r (3- 19) We assume C_sf = C_sr = C_s , s_f = s_r = s to simplify that equation. So we rewrite these equations.

2C_ss = −mγV_y (3- 20) −C_αf Vy+Lfγ Vx − δ − Cαr Vy−Lrγ Vx = m V y + γVx (3- 21) −C_αfL_f Vy+Lfγ Vx − δ + CαrLr Vy−Lrγ Vx = Izγ (3- 22) After finishing we obtain.

V _y γ = −Cα f+Cα r mVx −LfCα f+LrCα r mVx − Vx −LfCα f+LrCα r IzVx − Lf2Cα f+Lr2Cα r IzVx V_{γ +}y Cα f m LfCα f Iz δ (3- 23) V_y γ = 1 00 1 V_y γ + 00 δ (3- 24)

(41)

28

Second, we used the Dugoff tire model to calculate the road force exerted on the tire by the road.[23]

α_f =Vy+Lfγ Vx − θ2 (3- 25) s =−mγVy 2Cs (3- 26) V_s = V_x s2 _{+ tan}2_α _{(3- 27)} μ = μ₀ 1 − A_sV_s (3- 28) z = μFz 1−s 2 Css 2+ Cαtan α 2 (3- 29) if z < 1 f z = z 2 − z else f z = 1 (3- 30) F_x =Cssf z 1−s (3- 31) F_y = −Cαf z tan α 1−s (3- 32) F_r = F_xsinθ₂ + F_ycosθ₂ (3- 33) T_reaction = F_rR_p (3- 34)

Simplified model of steering rack and pinion system is Figure3.9.

We can change the model parameter to calculate external torque of vehicle. We assume vehicle on different road to test and verify the model. The

simulation result is Figure3.10. And we assume different vehicle velocity to test and verify the model. The simulation result is Figure3.11. From simulation results, we known the model can simulate the external torque by different road and vehicle velocity.

3.5 LQG controller

We have derived the LQ regular which has the deficiency of assuming all states are available for measurement, and we also have the development of the

(42)

29

Kalman filter, which produce an optimal estimation of the states. We can put them together and use the control law to obtain so-called LQG. The LQG control structure is Figure3.12. We used LQG controller to improve the SBW system response, made the system more stability.

(43)

30

Chapter 4 Simulation results

We used Table4.1, Table4.2, Table4.3 and equation to simulation. We used pole/zero map to analysis the systems. Figure4.1 is the pole/zero map of mechanical model. Figure4.2 is the red frame on pole/zero map of mechanical model. We know system response from the pole/zero maps. Figure4.3 is the pole/zero map of steering-by-wire model. Figure4.4 is the red frame on pole/zero map of steering-by-wire model. The SBW model is equivalent to mechanical model from their pole/zero maps. Figure4.5 is the pole/zero map of steering-by-wire model with LQG control. Figure4.6 is the red frame on pole/zero map of steering-by-wire model with LQG control. The reaction torque will change the K2 of model, so we change K2 to analysis pole/zeros changes as shown in Figure4.7 and Figure4.8.

The simulation of SBW model with LQG is Figure4.9 and Figure4.10. From the results of simulation we known the model with LQG control can filter out the measurement noise and made the model stability time to decrease.

We assume two cases to test and verify the response of SBW model with LQG better than SBW model.

4.1 Case 1: circular motion

Assume the vehicle drive on no wind and flat road, and the vehicle velocity is constant. One sec later we input the same torque human to the model to simulation the model on different road. And simulate the model of different vehicle velocity on the same road.

(44)

31

simulation of SBW model on different road accords with real life as shown in Figure4.11 and Figure4.12.

We knew the vehicle velocity faster and harder to steer. The simulation of SBW model of different vehicle velocity accords with real life as shown in Figure4.13 and Figure4.14. Than we compare the result of the model with LQG control and without LQG control on different road. We obtain the model with LQG control faster to stability than model without LQG control as shown in Figure4.15 and Figure4.16. We compare the result of the model with LQG control and without LQG control of different vehicle velocity. We obtain the model with LQG control faster to stability than model without LQG control as shown in Figure4.17 and Figure4.18. We know the model with LQG control to improve the stability apparent.

4.2 Case 2: turn a corner

Assume the vehicle drive on no wind and flat road, and the vehicle velocity is constant. One sec later we input the same torque human to steer a corner until 90 degree of the model to simulation the model on different road. And simulate the model of different vehicle velocity on the same road.

We knew the road friction coefficient larger and harder to steer. The simulation of SBW model on different road accords with real life as shown in Figure4.19 and Figure4.20. Because the road friction coefficient large and the vehicle locus outside.

We knew the vehicle velocity faster and harder to steer. The simulation of SBW model of different vehicle velocity accords with real life as shown in

(45)

32

Figure4.21 and Figure4.22. Than we compare the result of the model with LQG control and without LQG control on different road. We obtain the model with LQG control faster to stability than model without LQG control as shown in Figure4.23 and Figure4.24. We compare the result of the model with LQG control and without LQG control of different vehicle velocity. We obtain the model with LQG control faster to stability than model without LQG control as shown in Figure4.25 and Figure4.26. We know the model with LQG control to improve the stability apparent.

4.3 Case 3 : change line

Assume the vehicle drive on no wind and flat road, and the vehicle velocity is 20 km/h. One sec later we input the same torque human to change line on asphalt road.

We compare the result of the model with LQG control and without LQG control on this case. We obtain the model with LQG control faster to stability than model without LQG control as shown in Figure4.27 and Figure4.28. The longitudinal distance with LQG less than without LQG. The LQG control can improve steering response.

4.4 Conclusion

The SBW model can simulate the mechanical model equivalent. The SBW model without steer column can decrease the vehicle weight and the cost of cars. The tire model can simulate the external torque of vehicle on different road. Future we can use the tire model and steering wheel system of SBW model to simulate the steering wheel reactive torque of the vehicle on different road.

(46)

33

Reference

[1] 賴耿陽，“車輛驅動及控制＂，台南，復漢，1998。

[2] Hans B. Pacejka, “ Tyre and Vehicle Dynamics”, Butterworth-Heinemann, 2002.

[3] Bakker, E., Pacejka, H.B. and Lidner, L.,”A new tire model with an application in vehicle dynamics studies”, SAE Technical Paper Series, 890087, 1989.

[4] E Bakker, L Nyborg, H Pacejka, “Tyre modelling for use in vehicle dynamics studies”, Society of Automotive Engineers Paper, No.870421, 1987.

[5] M.米奇克，“汽車動力學 C 卷＂，北京，人民交通出版社，1997。 [6] Konghui Guo, Dang Lu, Shih-ken Chen, William C. Lin, Xiao-pei Lu,” The

UniTire model: a nonlinear and non-steady-state tyre model for vehicle dynamics simulation”, Vehicle System Dynamics, Jan, 2005.

[7] 郭孔輝，袁忠誠，盧蕩，“UniTire 輪胎穩態模型的聯合工況預測能力研究＂，汽車工程，6，NO.28，2006。

[8] Gillespie, T.D., “Fundamentals of Vehicle Dynamics”, Society of Automotive Engineers, 1992.

[9] Durstine, J.W., “The Truck Steering System from Hand Wheel to Road Wheel”, SAE SP-374, January 1974.

[10] Gillespie, T.D., “Front Brake Interaction with Heavy Vehicle Steering and Handling during Braking”, SAE Paper No. 760025, 1976.

(47)

34 NY, 1968.

[12] Taborek, J.J., “Mechanics of Vehicles”, Towmotor Corporation, Cleveland, OH, 957.

[13] Ned Mohan, Tore M. Undeland , William P. Robbins , ” Power Electronics: Converters, Applications, and Design”, John Wiley & Sons, Inc., 2003. [14] Reza N Jazar, “Vehicle Dynamics- Theory and Application”, Springer, 2008. [15] William F. Milliken , Douglas L. Milliken “Race Car Vehicle Dynamics”,

Society of Automotive Engineers, 1995.

[16] 吳秉霖，“力回饋方向盤於虛擬實境之發展＂，機械工程系所碩士論文，國立交通大學，2003. [17] 王柏堯，“車輛轉向系統方向盤力回饋控制技術研究＂，車輛工程研究所碩士論文，大葉大學，2006. [18] 林立璿，“車輛線控轉向系統研究與實作＂，車輛工程研究所碩士論文，大葉大學，2008. [19] 張碩編著，“自動控制系統＂，台北，全華，1998. [20] 清水浩，李添財編譯，“電動汽車全集＂，台北，全華，1997.

[21] Bimal K. Bose, Modern Power Electronic and AC Drives, Printice-Hall, 2002.

[22] Ji-Hoon Kim and Jae-Bok Song, “Control logic for an electric power steering system using assist motor,” Science Direct journal, Mechatronics, pp.447-459, April 2002.

[23] Dugoff, H., Fancher, P.S., Segel, L., "An Analysis of Tire Traction Properties and Their Influence on Vehicle dynamic Performance," SAE Paper, No.700377, 1977

(48)

35

[24] R.W. Rivers, Technical traffic accident investigators' handbook: a level 3 reference, training, and investigation manual, Thomas, 1997.

[25] Yu Lei-yan, “Research on control strategy and bench test of automobile Steer-by-Wire system”, IEEE Vehicle Power and Propulsion Conference (VPPC), September 3-5, 2008.

[26] CJ Kim, , J-H Jang, S-N Yu, S-H Lee, C-S Han, J K Hedrick,” Development of a control algorithm for a tie-rod-actuating steer-by-wire system”, Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, Volume 222, Number 9 / 2008.

(49)

36

Figures

Figure1.1 The physical model of a tire

(50)

37

Figure1.3 Schematic block diagram of the system

(51)

38

Figure2.2 Steer rotation geometry at the road wheel

(52)

39

Figure2.4 Front view of a tire and measurement of the kingpin incline angle

(53)

40

Figure2.6 A positive and negative caster configuration on front wheel of a car

(54)

41

Figure2.8 SAE tire force and moment axis system

(55)

42

Figure2.10 Moment produced by vertical force acting on K.P.I. angle

(56)

43

Figure2.12 Steering moment produced by traction force

(57)

44

Figure2.14 Steering linkages model

(58)

45

Figure3.1 Mechanism diagram of mechanical steering system

Figure3.2 Math model of the mechanical steering system K₂ K B B₂ J₂ J₁ θ1 θ₂ B₁ ≅ 0 _K₁ _{≅ 0} T_hum Text T_ext T_hum

(59)

46

Figure3.3 Steering wheel system and front wheel system

Figure3.4 Math model of steering system K₂ B₂ J₂ J₁ θ₂ θ₁ B₁ ≅ 0 _K₁ _{≅ 0} T_hum T_ext T₁ T₂ T_ext T₂ T_hum T₁ Steering wheel system Front wheel system

(60)

47

Figure3.5 Steer by wire system T_ext T₂ T_hum T₁ M M

(61)

48

Figure3.6 Control block diagram of SBW

1 L1s + R1 1 L₂s + R₂ k_τ1 k_τ2 kϕ1 k_ϕ2 1 J ₁s 1 J₂s 1 s 1 s B1 B2 K2 n K + Bs Q K + Bs T_hum T_ext T1 T2 ω1 θ1 θ₂ ω2 + + − + − + + − − + + + + + − ₋

(62)

49

Figure3.7 SAE vehicle axis systems. [8]

(63)

50

Figure3.9 Simplified model of steering system [22]

Figure3.10 Reaction torque on different road

0 5 10 15 20 25 30 0 2 4 6 8 10 12 14 16 18 20 Torque-ext, V=20 Km/h

front wheel angle (degree)

To rq u e ( N -m ) ₀ = 0.8 ₀ = 0.6 ₀ = 0.4 ₀ = 0.2

(64)

51

Figure3.11 Reaction torque of different vehicle velocity

Figure3.12 LQG control structure

0 5 10 15 20 25 30 0 5 10 15 20 25 Torque-ext, ₀ = 0.8

front wheel angle(degree)

To rq u e ( N -m ) V = 10 km/h V = 20 km/h V = 30 km/h V = 40 km/h u(n) x n + 1 = Ax n + Bu n + Gw n y n = Cx n + Du n + Hw n Steer-by-wire system LQG regulator Kalman estimator -K w(n) T_hum, T_ext y(n) θ₁, θ₂ x (n) ₊ + v(n) y_v(n)

(65)

52

Figure4.1 Pole/Zero map of mechanical model

Figure4.2 Red frame on mechanical model

Pole-Zero Map Real Axis Im a g in a ry A x is -1 -0.5 0 0.5 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T Pole-Zero Map Real Axis Im a g in a ry A x is 0.8 0.85 0.9 0.95 1 1.05 -0.1 -0.05 0 0.05 0.1 0.1/T 0.1/T 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 System: dsys_tr Pole : 0.993 Damping: 1 Overshoot (%): 0 Frequency (rad/sec): 0.719

(66)

53

Figure4.3 Pole/Zero map of steer-by-wire models

Figure4.4 Red frame on steer-by-wire models

-1 -0.5 0 0.5 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T Pole-Zero Map Real Axis Im a g in a ry A x is 0.8 0.85 0.9 0.95 1 1.05 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.7 0.8 0.9 System: dsys_sbw Pole : 0.993 Damping: 1 Overshoot (%): 0 Frequency (rad/sec): 0.687 Pole-Zero Map Real Axis Im a g in a ry A x is

(67)

54

Figure4.5 Pole/Zero map of SBW model with LQG control

Figure4.6 Red frame on SBW model with LQG control

-1 -0.5 0 0.5 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.1/T 0.2/T 0.3/T 0.4/T 0.5/T 0.6/T 0.7/T 0.8/T 0.9/T /T Pole-Zero Map Real Axis Im a g in a ry A x is Pole-Zero Map Real Axis Im a g in a ry A x is 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 System: dsys_lqg Pole : 0.982 + 0.0179i Damping: 0.707 Overshoot (%): 4.32 Frequency (rad/sec): 2.58

(68)

55

Figure4.7 SBW model root locus

Figure4.8 SBW model with LQG root locus

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 Pole-Zero Map Real Axis Im a g in a ry A x is Pole-Zero Map Real Axis Im a g in a ry A x is -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1

(69)

56

Figure4.9 Simulation with LQG, when input is torque human

0 10 20 30 40 50 60 70 80 90 100 -2 -1 0 1 2 T o rq u e (N -m ) Time (sec) T

input and Treaction

0 10 20 30 40 50 60 70 80 90 100-20 0 20 40 60 T o rq u e (N -m ) 0 10 20 30 40 50 60 70 80 90 100 -50 0 50 100 150

steering wheel angle

time (s) a n g le ( d e g re e ) 0 10 20 30 40 50 60 70 80 90 100 -2 0 2 4 6

front wheel angle

time (s) a n g le ( d e g re e )  1+v(n):without LQG  1+v(n):with LQG  1 estimation ₂+v(n):without LQG  2+v(n):with LQG  2 estimation

(70)

57

Figure4.10 Simulation with LQG, when input is external torque

0 10 20 30 40 50 60 70 80 90 100 -20 -10 0 10 20 T o rq u e (N -m ) Time (sec) T

input and Treaction

0 10 20 30 40 50 60 70 80 90 100-20 0 20 40 60 T o rq u e (N -m ) 0 10 20 30 40 50 60 70 80 90 100 -50 0 50 100 150  1 time (s) a n g le ( d e g re e ) 1+v(n):without LQG  1+v(n):with LQG ₁ estimation 0 10 20 30 40 50 60 70 80 90 100 -2 0 2 4 6  2 time (s) a n g le ( d e g re e ) 2+v(n):without LQG  2+v(n):with LQG  2 estimation

(71)

58

Figure4.11 Case 1: simulation on different road

Figure4.12 Case1: vehicle locus on different road

0 10 20 30 40 50 0 0.5 1 1.5 2 2.5 T_hum, V=20km/h time (s) to rq u e ( N -m ) ₀=0.8 ₀=0.6 ₀=0.4  0=0.2 0 10 20 30 40 50 -20 -15 -10 -5 0 5 T_reaction, V=20km/h time (s) to rq u e ( N -m ) 0 10 20 30 40 50 -200 0 200 400 600 Time (sec) a n g le ( d e g re e ) steering wheelangle 0 10 20 30 40 50 -10 0 10 20 30 Time (sec) a n g le ( d e g re e ) front wheelangle -10 0 10 20 30 40 50 60 -10 0 10 20 30 40 50 60 Vehicle locus, V = 20 km/h X (m) Y ( m ) ₀=0.8 ₀=0.6 ₀=0.4 ₀=0.2

(72)

59

Figure4.13 Case 1: simulation of different vehicle velocity

Figure4.14 Case1: vehicle locus of different vehicle velocity

0 10 20 30 40 50 0 1 2 3 T hum, 0=0.8 time (s) to rq u e ( N -m ) V=10km/h V=20km/h V=30km/h V=40km/h 0 10 20 30 40 50 -30 -20 -10 0 10 T reaction time (s) to rq u e ( N -m ) 0 10 20 30 40 50 -200 0 200 400 600 Time (sec) a n g le ( d e g re e ) steering wheelangle 0 10 20 30 40 50 -10 0 10 20 30 Time (sec) a n g le ( d e g re e ) front wheelangle -10 0 10 20 30 40 50 60 70 -10 0 10 20 30 40 50 60 70 Vehicle locus, ₀=0.8 X (m) Y ( m ) V=10km/h V=20km/h V=30km/h V=40km/h

(73)

60

Figure4.15 Case 1: simulation with and without LQG control

Figure4.16 Case 1: vehicle locus with and without LQG control

0 10 20 30 40 50 0 0.5 1 1.5 2 2.5 T_hum, V=20km/h time (s) to rq u e ( N -m ) ₀=0.8 ₀=0.6 ₀=0.4 ₀=0.2 ₀=0.8 with LQG ₀=0.6 with LQG ₀=0.4 with LQG ₀=0.2 with LQG 0 10 20 30 40 50 -20 -15 -10 -5 0 5 T_reaction, V=20km/h time (s) to rq u e ( N -m ) 0 10 20 30 40 50 -100 0 100 200 300 400 500 600 Time (sec) a n g le ( d e g re e ) steering wheelangle 0 10 20 30 40 50 -5 0 5 10 15 20 25 30 Time (sec) a n g le ( d e g re e ) front wheelangle -20 -10 0 10 20 30 40 50 60 -10 0 10 20 30 40 50 60 Vehicle locus, V = 20 km/h X (m) Y ( m )  0=0.8 ₀=0.6 ₀=0.4  0=0.2  0=0.8 with LQG ₀=0.6 with LQG ₀=0.4 with LQG  0=0.2 with LQG

(74)

61

Figure4.17 Case 1: Simulation with and without LQG control

Figure4.18 Case 1: vehicle locus with and without LQG control

0 10 20 30 40 50 0 0.5 1 1.5 2 2.5 3 3.5 T_hum, ₀=0.8 time (s) to rq u e ( N -m ) 0 10 20 30 40 50 -25 -20 -15 -10 -5 0 5 T_reaction time (s) to rq u e ( N -m ) 0 10 20 30 40 50 -100 0 100 200 300 400 500 600 Time (sec) a n g le ( d e g re e ) steering wheelangle 0 10 20 30 40 50 -5 0 5 10 15 20 25 30 Time (sec) a n g le ( d e g re e ) front wheelangle V=10km/h V=20km/h V=30km/h V=40km/h V=10km/h with LQG V=20km/h with LQG V=30km/h with LQG V=40km/h with LQG -20 -10 0 10 20 30 40 50 60 70 -10 0 10 20 30 40 50 60 70 Vehicle locus, ₀=0.8 X (m) Y ( m ) V=10km/h V=20km/h V=30km/h V=40km/h V=10km/h with LQG V=20km/h with LQG V=30km/h with LQG V=40km/h with LQG

(75)

62

Figure4.19 Case 2: simulation on different road

Figure4.20 Case 2: vehicle locus on different road

0 5 10 15 20 25 -2 -1 0 1 2 2.5 To rq u e ( N -m ) Time (sec) T_hum and T_reaction, V=20 km/h

0 5 10 15 20 25-20 0 20 25 To rq u e ( N -m ) 0 5 10 15 20 25 -100 0 100 200 300 400 ₁ time (sec) a n g le ( d e g re e ) 0 5 10 15 20 25 -30 -20 -10 0 10 20 30 2 a n g le ( d e g re e ) Time (sec) ₂ and  0 5 10 15 20 25-5 5 15 25 35 45 55  a n g le ( d e g re e ) 0 5 10 15 20 25-5 5 15 25 35 45 55 0 5 10 15 20 25-5 5 15 25 35 45 55 0 5 10 15 20 25-5 5 15 25 35 45 55 0 5 10 15 20 25 -5 0 5 10 15 direction angle (₂ - ) time (sec) a n g le ( d e g re e ) 0 10 20 30 40 50 60 70 80 -10 0 10 20 30 40 50 60 70 Vehicle locus, V = 20 km/h X (m) Y ( m )

車輛方向盤於不同路面操控模擬

國 立 交 通 大 學

機械工程學系

碩 士 論 文

車輛方向盤於不同路面操控模擬

Simulation on Different Road Surface of Vehicle Steering

Wheel

研 究 生：林志隆

指導教授：成維華 教授

車輛方向盤於不同路面操控模擬

車輛方向盤於不同路面操控模擬

摘要

Simulation on Different Road Surface of Vehicle Steering Wheel

Abstract

誌 謝

Contents

List of Tables

List of Figures

Chapter 1 Introduction

1.1.Motive and objectives

1.2. Literature Review

1.2.1. Faila’s model





































β

1.2.2. Magic Formula

   

 





 





1.2.3. Mitschke’s model













sin 2

sin

2

Q D

M













1.2.4. UniTire model























 

國立交通大學

碩士論文

研究生：林志隆

指導教授：成維華教授

誌謝