Designing of the Hybrid Adaptive Cruise Control-Stop&GO

bigskymanAI and Robotics

Oct 24, 2013 (4 years and 16 days ago)

96 views

UKACC Research Student Presentations, 7 May 2009



1

1



Abstract


Over the recent year, a considerable growth
in

the
number of vehicle
s

on the road, has been observed. This
increases importance of vehicle safety a
nd m
inimization of fuel
consumption,

subsequently

prompting manufacturers of cars
to equip their product
s
, with more advanced features such as
Adaptive Cruise Control (ACC), Collision Avoidance and
Collision
W
arning
S
ystem (CWS).

In this PhD project we
c
on
centrate on new methods for ACC,

for optimization of fuel

consumption in urban traffic. T
his work will include:

D
esign of
the simulation model suitable for this application
,

d
esign of
virtual traffic environment

bas
ed on the models
,

i
nvestigation
and

desig
n of suitable
h
ybrid control algorithm,

most likely
using the ideas of non
-
linear predictive control
,

r
eal
-
time
im
plementation of the algorithms and

tes
t
s
.


I.

I
NTRODUCTION


daptive cruise control (ACC) is the

extension

of the
Cruise Control system (CC
),

it is
able to vary the
velocity

of the vehicle

depending

o
n

the
behavior

of the
other vehicles moving in front of the vehicle equipped with
ACC by applying the brake and modulating the throttle
to
produce the necessary power
.
This

system uses

the
radar
or

other sensory devices

to measure the
distance
between

vehicles (
the time
takes to
return the beam to
emitter
)
(
figure.
1
)
. [
1
]

The extended version of the ACC is so
-
called ACC
stop&go
.

In the urban area and traffic jam
situation

where
driver needs freq
uently to apply the accelerator

and
brake

pedal the

stop&go system

make
s

the driving less tire
-
some
.

Unlike the ACC
normally working

for

high
-
speed
s

this
system works in the speed range of less than 30km/h and
slowing down to zero.

F
igure.
2

classified the
characteristics

and
maximum braking capacity for ACC and Stop&Go
system.


It is known that ACC is capable of managing

the traffic
flow,

b
y
making platoons of vehicles
it

improve
s

highway
capacity.
Under

this condition a lot of vehicles moving at
highway sp
eed behind other ones

with small inter
-
distance
can

increase density of the vehicles in the
highway.
I
t also
has

the positive effect on the optimization of fuel
consumption
e
specially for

heavy

vehicle
s this is

due to
significant effect of the
a
erodynamic drag
dependent on the
cross
section
front area

for such vehicle
.

[2
]

Another version of the velocity

controlling syste
m has been
introduced

along with de
veloping the CC and ACC system
which uses the information
about the r
oad ahead
of the
vehicl
e
to meet the aim of the reducing the fuel consumption

which is so
-
called Look Ahead Cruise
Controller [
3
]
.

I
n
this


case some derivative velocity control
ling system is

introduced
as instance

Predictive Cruise
Control (
PCC)

[
4
],
Expert Cruise Control

(ECC
) [
5
] and Model Predictive
control (
MPC
) [
6
].





Figure.1 Performance of the ACC system
overview

[14
]





Figure.
2

differences

between
ACC and Stop&Go system [15]




Figur.
3

a truck using Look A
head

Cruise
Controller [
3
]



II.

VEHICLE
M
ODEL


To

design

an
ACC
system and
to
carry

out the investigation
on it
,

the

longitudinal
dynamic
model of the vehicle is
essential.


Designing of the Hybrid Adaptive Cruise
Control
-
Stop&GO

Payman Shakouri
,
Prof. And
rz
ej Ordys,
Dr. Mohamad Askari


Faculty of Engineering
-
Kingston University






A

UKACC Research Student Presentations, 7 May 2009



2

2


A.


Longitudinal model of the vehicle


Using
the second

Newton’s
law, the

relationship between
resis
ting
force (
Rolling resistance, A
ero
dynamic
drag)
,
brake

torque and wheel t
orque

obtained from figure.
4

and
defined by following equation:




̇



(






)









(1)


Where



is the wheel torque,




is the brake torque,



is related to aerody
namic Force and last terms in
equation.1 defines the
rolling resistance and gravitational
forces respectively.


As Aerodynamic drag force is a function of the square of the
speed thus is the most significant parameter
affecting on the
performance of the vehicle
in velocity of more than 48 km/h
[
7
],
it can

be determined
as following function:










(




)




(
2
)


Where


is the air density

which
equals to

1.225 kg/m
3
, A
is
maximum

vehicle cross area

in [m
2
]. I
n [
7
] an approximated
relationship for this parameter was determined as function
of the car mass





in the range of 800
-
2000
kg:















(





)







(
3
)







is the drag Coefficient

varying for different body
shape
,


is velocity of the vehicle

(m/s
2
)
and



air veloc
ity

which
could be neglected in the calculations.

The friction coefficient is nonlinear function of the
slip

(=s)
,





(

)
.W
ithout

braking the vehicle
velocity is
almost
the
same as the wheel
velocity (


0





)
.

But in [
8
]

friction coefficient consider as constant and equal to 0.5 in
norma
l road condition.
S
lip

is

determined

as follows
:


slip =1
-






(4)












: Vehicle linear
velocity


:
Wheel radius



:
wheel angular velocity

B.

Powertrain


Automatic transmission to transfer the torque from engine to
gearbox applies a torque convertor.
The t
orque convertor
consists of three essential parts including impeller, the
turbine and
the reactor. Impeller is connected to crankshaft
which transmits the power t
o
turbine by the hydraulic oil
inside of the torque
convertor.
In turn the
turbine is
connected to the output shaft of the convertor which is
coupled to the input shaft of the gear
box. The torque v
aries
depending upon gear ratio
. Finally the
output
torque from
the gearbox will be transmitted to the wheels after passing
through the final drive
.

The modeling of
the
powertrain

follows the method being
applied in [
9
].





Figure.
4

schematic model of the vehicle showing various


f
orces acting on it during moving on the road






Figure.5
Transmission of the torque and velocity form engine to wheels


The equation
determining the difference between engine
torque and impeller torque

i
s defined: [
9
]









̇







(
5
)

Where;



: Engine speed

(rpm)




: Summation of engine and impeller moment of inertia

(Kg.m
2
)



: Engine torque

(N.m)



: Impeller torque

(N.m)


The schema
tic model of the vehicle
powert
ra
i
n

is depicted
in Figure.
5
.

The parameters pl
aying the important role on the
performance of a torque convertor are expressed as
following [
12
]:

Speed ratio=


=



/







Torque ratio
=



=



/



Efficiency =









Capacity factor or K
-
factor =


= speed /




The capacity factor shows the ability of the convertor to
absorb or

transmit torque [
7
].

F
igure.
6

illustrate
s

the
relationship between the above parameter in which the
torque ratio, efficiency and
input capacity factor are plotted
against the speed ration [
7
].
Input
torqu
e of

the convertor
can be define as follows
:






(






)


(
6
)


UKACC Research Student Presentations, 7 May 2009



3

3




Figure.
6

performance characteristic of a torque convertor [12]



Figure.
7

Capacity factor of an internal combustion engine [
12
]



Where





the impeller torque (
converter input torque
)

and



is the impeller speed (
converter input speed
)
.

The

engine and the convertor should have a similar rang
e

of
capacity factor
s

(





)

to achieve proper matching.


Having the

Engine operating point
,






can be obtained
from figure.
7
.

Interpolating

from

figure.
6

the speed
ratio




will be found.

The

speed
ratio



,

the torque ratio




of the torque
convertor and



can be determined as function of the







.

Because






and

the velocity of the impeller






equals to the engine speed




therefore the torque of the
impeller can be found
. So

that the equation (
6
) can be
modified as following:






(






)


(
7
)


Having the



,




enable us to find the output
characteristic of the torque convertor:

















(
8
)
















(
9
)





: Turbine torque

N
t :

Turbine angular velocity










(




)
=

(




)

Speed ratio
=





(




)
=

(




)


Knowing that the torque and speed output from the torque
convertor equals to the input characteristic of the gear
box

help us to find the outpu
ts of the gearbox:




Figure.8
Block diagram of the vehicle














(
10
)












(
11
)


Where




















are
respectively
transmission input

and
output
torque
,
transmission
output
and

input speed.



is transmission ratio

and
varies
depending on shifting
gear number, in another

wo
rd is
a
function of the gear setting.

Multiplying the transmission output torque by final drive
ratio determines the wheel
torque,




which can be used to
obtain the angular velocity of the wheel.

It is assumed that there is
a single source actuation system
due to the fact that
under normal
operation, the

vehicle
deceleration is in range
of
-
2m/s
2
,

so
the rear and the front
brakes can be expected to

have same brake
pressure [
12
]

therefore
the total

brake torque can be used
in

calculation. I
n
reality
while braking the load transfers from rear axle to
front axle [
7
] that in turn request the different braking force
and also the
maximum braking force must be introduced for
each axle separately in order to preventing the vehicle from
slipping and guarantee the safety of the vehicle, that is the
reasons of
the
innovation of the ABS (Anti Brake System
).

Las
t

term in equ
ation (1)

is gravitational force
acting

on
vehicle
travelling

uphill
or
downhill and sign of this force
would be negative or positive accordingly.

Consequently

̇

wh
ich is
the
acceleration

of the vehicle is
obtained
.

By integrating

of
this

parameter
the velocity of

the
vehicle
will
be obtained.

Figure.8 depicts the block diagram of the vehicle in
simulation.


III.

CONTROLLER


In designing the controller system the control objectives
should be identified and suitable criteria and constrains on
the base of the specified t
argets is devised.

The typical controller method is
the
PID controller which
has
been affluently used by

the industry. To guarantee the
robustness and reducing disturbances in the
system, the

new
method such as fuzzy logic and Model
Predictive

Control
(
MPC
)
, etc has been introduced.

Linear system can
consistently work with existing control
method
and to deal with any non
-
linearity

in the system
the
linearization
methods could
be applied to
approximate

th
e
nonlinear system to linear

and then current linear control
techniques could easily be used for such

the

systems
. [
12
]

UKACC Research Student Presentations, 7 May 2009



4

4

Actually the model of ACC
consists

of two control
loops.

Inner
-

loop
consisting
of CC

and model of the vehicle
works
as tracking velocity and try
ing to reduc
e the q
uantity of the
error (difference between actual velocity and desired
velocity)

so the inner
-
loop introduced as CC system. Outer
loop

works

as tracking vehicle
system
and calculates the
new desired value of velocity being applied as
reference

value for the
inner
-
loop. The

outer loop by using the input
data including

the actual velocity of the vehicle, distance
from leading vehicle, desired velocity and desired distance
calculates the new value of the velocity for the CC.
[12]
(figure.8)


IV.

VIRTUAL TRAFFIC ENVI
RONMENT

To guarantee
the functionality of

the ACC simulation model
over the various different traffic situations, the
traffic
model
is needed to
represent
different
traffic
condition
s
.ASM
Traffic

which has been developed by “
dspace


[1
3
] provides

us
with
reliable, flexible and time
-
saving testing facility by
mean
s of creating any kind of
traffic scenario.

Some traffic scenario
s where a

vehicle equip
ped with ACC
could
encounter

in real
ity are as follows:




When the Vehicle equipped with ACC is faster
than t
he leading vehicle;




More than one vehicle drive on the different lane
from the vehicle fitted with ACC
and might one of
them cuts in the following vehicle’s lane;




Leading and following vehicle travel at the same
speed and with constant distance from each

other,
suddenly the leading vehicle leaves the lane.




In oncoming traffic when the
vehicle equipped with
ACC drive at higher speed than vehicle
ahead and
have to reduced the velocity until an oncoming
vehicle has passed to perform a lane change




A
pedestrian cross
es

the road

where
the fully
stop
of

th
e vehicle or swerving by driver is needed.


V.

C
ONCLUSION
S

In this paper w
e explain briefly about Automatic driver
assistance system which

reduce the risk of accident, provide
the
driver with comfort and s
afety and reduce
fuel
consumption. All of these devices uses the same concept but
with different point of view which can be found as their
i
mpact of the safety and comfort. The model of the vehicle
was explained and formulated. The block diagram of the
ACC

was expressed. Finally the brief explanation of the
virtual traffic envir
onment was presented. In the PhD

project
a

preci
se

study and verification will
be

carried out on
all
mentioned

issues

and new method
s

will be
introduced
.








Figure.8
the
block diagram of the ACC (inner
-
loop & outer
-
loop)

[
12
]





Figure.9 3
-
D Visualization of the ASM Traffic in MotionDesk

[dspace]

A
CKNOWLEDGEMENT

I would like to appreciate of my supervisor
y team Pro. A.
Ordys and Dr. M.
Askari for their valuable
help. I’m

also
grateful of my parents who without their support get g
oing
on this way would be impossible.

R
EFERENCES

[1]

H.Winner, K.Winter, B.Lucas et al,

ACC Adaptive Cruise Control
,The Bosch Yellow Jacket,2003

[2]

A.Vahidi, A. Eskandarian, Research Advances in
Intelligent Collision
Avoidance and Adaptive Cruise Control,

IEEE Trans. Intelligent
Transportation system
, vol. 4, September 2003.

[3]

Look
-
Ahead Control for heavy truck to minimize trip time and fuel
consumption, E.

Hellström,

M.

Iva
rsson, J. Åslund, L. Niel
sen,
ScienceDirect, 2008

[4]

F.
Lattemann. et al,The
Predictive

Cruise Control
-
A System to reduce
Fuel Consumption of Heavy Duty
Trucks, SAE

paper,2004
-
01
-
2616

[5]

A.Wingren,Fordonsreglering med F
ramforhallninn ,MSc Thesis
Linkö
ping University

[6]


D.Axehill
,J.

Sjöberg
,
Adaptive

Cruise Control for Heavy Vehicles

Hybrid Control and MPC
A,Linkopings University,2003


[7]

J. Y. Wang, Theory of Ground Vehicle, third Edition, p.265
-
268,
2001

[8]

S. C. T. Filho, D. C. Donha, Automobile Stop
-
AND
-
Go Cruise
Control System Tuned b
y Genetic Algorithm, University of Sao
Paulo, 2004

[9]

U
sing simulink and Stateflow in Automotive Application

[10]

S. C. T. Filho, D. C. Donha, Automobile Stop
-
AND
-
Go Cruise
Control System Tuned by Genetic Algorithm, University of Sao
Paulo, 2004

[11]

D. B. Maciuca, Brake
modeling

and Control, p.409
-
410, L.Vlacic,
M.Parent, F.Harashima, Intelligent Vehicle Technologies

[12]

P.Riis, Adaptive Cruise Controller Simulation as an Embedded
Distributed
System,

MSc thesis, Linkoping University, 2007

[13]

dspace Gmbh ,
ASM

Traffic
,2008

[14]

www.audiusa.com/.../safety_technology.html


[15]

D.Maurel, AAC System
-
Overview and examples
, Brake modeling
and Control, p.4
2
9
-
, L.Vlacic, M.Parent, F.
Harashima, Intelligent
Vehicle Technologies