4

Vehicle Dynamic Control of 4 In-Wheel-Motor

Drived Electric Vehicle
Lu Xiong and Zhuoping Yu
Tongji University
China

1. Introduction
Thanks to the development of electric motors and batteries, the performance of EV is greatly
improved in the past few years. The most distinct advantage of an EV is the quick and
precise torque response of the electric motors. A further merit of a 4 in-wheel-motor drived
electric vehicle (4WD EV) is that, the driving/braking torque of each wheel is independently
adjustable due to small but powerful motors, which can be housed in vehicle wheel
assemblies. Besides, important information including wheel angular velocity and torque can
be achieved much easier by measuring the electric current passing through the motor. Based
on these remarkable advantages, a couple of advanced motion controllers are developed, in
order to improve the handling and stability of a 4WD EV.

2. Traction control
The fast and accurate torque generation of each driving wheel enables a great enhancement
in traction control during acceleration.
In this section, an anti-slip controller for a 4WD EV using VSC (Variable Structure Control)
method is presented. The control algorithm is independent on the identification of the road
adhesion coefficient and has excellent robustness to the estimation error of the vehicle velocity.
Regarding the high-frequency-chattering on the sliding surface, a new control method which
combines the advantage of the VSC and MFC (Model Following Control) in order to decrease
the fluctuation to the e-motor torque and slip ratio of the tire is proposed. The result of the
simulation indicates that the proposed control method is effective for the ASR control and
improves the performance of e-motor’s output torque and the slip ratio of the tire.

2.1 VSC ASR controller

2.1.1 One-wheel-model
An accurate simulation model is important to verify the effect of the designed controller.
Fig.2.1-1 shows a two degrees of freedom vehicle model. It only contains the vehicle’s
longitudinal motion and ignores air resistance and rotating resistance. Formula 2.1-1 shows
the mathematical model:


M v x  Fd

www.intechopen.com
68 Electric Vehicles  Modelling and Simulations


I w   Tm  Fx R (2.1-1)

Here, M is the 1/4 vehicle mass, kg; vx represents the longitudinal velocity, m/s; Fx is the
driving force of the road, N; Iω is the wheel rotational inertia, kg·m2; R is the wheel radius,
m; ω is the angular velocity, rad/s and Tm is the motor torque, N·m.
The Magic Formula tire model is applied as the tire model, so the driving force Fd can be
expressed as follows:


Fd   Max  FZ  sin C  arctan B  1  E    E  arctan( B )  (2.1-2)

The meanings of the parameters can be found in the literature [1].

Fig. 2.1-1. One-wheel-Model

2.1.2 Design of VSC ASR controller

VSC with sliding mode has good robustness to the input signals so that this strategy has
advantage to the ASR control system which needs the vehicle velocity observation and
signal identification. But there is always high-frequency-chattering on the sliding surface. In
the following text a VSC controller, which doesn’t depend on the identification of the
optimal slip ratio, is designed and its performance will be analyzed through simulation.
In order to make the VSC possess excellent robustness to the additional uncertainties and
interferences, the control law adopted here is equivalent control with switching control.
Hence, the output torque of the e-motors can be expressed as [2]:

Tm  Tm , eq  T sgn(s ) (2.1-3)

In this equation, Tm,eq is the equivalent torque of the e-motor, ΔT is the hitting control drive
torque, sgn(s) is the switching function of the system.
The sliding motion includes two processes: approaching motion and sliding motion. The
approaching motion can make the system at any time in any position approach to the
sliding face in limited time. The sliding motion occurs only when the system reaches sliding
surface:

s    reference  0 .

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 69

Fig. 2.1-2. Diagram of VSC ASR Strategy

reference slip is constant, so  reference  0

To reach the ideal sliding mode,.
the requirement s=0 should be fulfilled. Assuming the

So, on the sliding face there is:

 
   reference  0 (2.1-4)

According to the one-wheel model:


I   Tm  Fx R

During driving process, the slip ratio of the wheel can be expressed as:

R  v

R
Combining Formula (2.1-1) and (2.1-4), we can get:

d 1  Tm  Fx R 
  v  (1   )R 0
dt R  I 
Then, we can obtain the e-motor’s equivalent torque:


Tm , eq  v  Fx R
I
R(1   )

As the tire’s longitudinal velocity is difficult to be measured accurately, v is the estimated
value. Then the above formula can be rewritten as:


 
T m , eq  v  Fx R
I
R(1   )
(2.1-5)

In the actual driving progress, there are many kinds of road surfaces and their respective
optimal slip ratios. The identification for them is difficult. Through Fig. 2.1-3, we can see that,
although the slip ratios for different roads are different, the basic shapes for - curves are

www.intechopen.com
70 Electric Vehicles  Modelling and Simulations

similar. It means, before the optimal slip ratio, the bigger the slip ratio, the larger the
longitudinal adhesion coefficient is. While after the optimal slip ratio, the bigger the slip
ratio, the smaller the longitudinal adhesion coefficient is[3].

Fig. 2.1-3. Slip ratio-Longitudinal adhesion coefficient on different road surface

d
From Fig. 2.1-3, we can get:
 0 ,   reference ,  needs increasing so as to get larger adhesion coefficient and
d
When

d
the driving torque should be increased.
 0 ,   reference ,  needs keeping so as to get larger adhesion coefficient and the
d
When

d
driving torque should be maintained.
 0 ,   reference ,  needs decreasing so as to get larger adhesion coefficient and
d
When

the driving torque should be reduced.


T I 
According to the one-wheel model, we can acquire:   m 
FZ R
Then we can get:

 
Tm  I   
d d / dt  2 Tm  I 
  
d d / dt
FZ R
 
. 
vR  v R v  v
FZ

( R )2

Now, we can get the judgment condition:

 
Tm  I 
 
 0 , the e-motor’s output torque needs increasing;
v  v
When

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 71

 
Tm  I 
 
 0 , the e-motor’s output torque needs keeping;
v  v
When

 
Tm  I 
 
 0 , the e-motor’s output torque needs decreasing.
v  v
When

From above we can find that what the switching function needs is not the slip ratio and the
reference slip ratio any more, but the angular speed, e-motor’s torque and driving torque,
which need not identification. Although there is still longitudinal velocity estimation value
in the controller, the controller itself has solved this problem which can be seen in Formula 8.
So this VSC strategy is considered as feasible.
When the system is not on the sliding surface, it needs approaching the sliding surface from
any state. This motion is called approaching motion. And during this motion the slip ratio
will be approaching 0.
Under the generalized sliding condition, the switching function should meet:


s s   s (2.1-6)

Here the parameter  >0.  represents the velocity, in which the system approaches the
sliding surface. The larger the  is, the faster the approaching velocity is. Whereas, the
chattering on the sliding surface will be bigger.
When Formula (2.1-1) is put into Formula (2.1-6), we can get:

s  T  T sgn(s )  Fx R
 [ v  (1   )R m ]   s
R
(2.1-7)
I
Here the hitting control driving torque is assumed as

 I
T  (F   )
(1   )
(2.1-8)

Putting Formula 2.1-7 into Formula 2.1-6, we can get:


1  
( vx  vx ) s  F s
R

That is:


1  
F vx  vx
R
(2.1-9)

So the e-motor’s output torque can be shown as


Tm  T m , eq  T sgn( s ) (2.1-10)

www.intechopen.com
72 Electric Vehicles  Modelling and Simulations

The simulation results for vehicle that starts on the road surface with a low adhesion
coefficient =0.2 is shown in Fig.2.1-4.

Fig. 2.1-4. Start on a low adhesion surface =0.2

From the simulation results we can get that the vehicle can keep away from skipping and
the acceleration performance is good when it starts. But the slip ratio occurs fluctuation
when it’s among 0 to 0.3 and the e-motor’s output torque also fluctuates near 300Nm. In
reality, big fluctuation is harmful to the e-motor and sometimes the e-motor can’t fully
realize what the controller requires. Therefore, there are some defects in this method.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 73

2.2 ASR combined controller

2.2.1 MFC controller
According to the research results from Tokyo University [4, 5], when the tire is completely
adhered, the vehicle’s equivalent mass is equal to the sum of the sprung mass and non-
sprung mass. When the tire slips, the angular speed changes significantly. During
acceleration, the angular speed is obviously smaller than the ideal value which is outputed
by the standard model. In light of this principal, the tire’s angular speed should be
compared to the angular speed from the standard model at any time. And then the
difference is used as the basis for a correction value through a simple proportional control to
adjust the e-motor’s output torque. So that the tire can avoid slipping.
MFC strategy only requires the e-motor’s output driving torque and the tire’s angular speed
signal to put ASR into practice. Consequently the estimation of the longitudinal velocity and
the optimal slip ratio identification can be ignored. Therefore, this strategy is practical. The
system diagram is shown in Fig.2.2-1.

Fd

F M + FM + - 1 Vw
mw s
-
dF

1
( m  mw ) s
- +

 m  mw  s
 s 1
1

Fig. 2.2-1. MFC control block diagram

The standard model of MFC is got under the condition that the slip ratio is set to 0. It means
that the road’s adhesion force isn’t fully utilized and the driving performance will be bad. So
this control strategy is not perfect. Secondly, MFC hasn’t good robustness to the input
signals. Especially when the angular speed is disturbed, deviation of the controller will
happen.

2.2.2 Combined controller

Based on the characters of VSC and MFC, in this section an area near the sliding surface will
be set, within which the MFC strategy is applied. And out of this area, the VSC strategy is
used.Thus, the high-frequency-chattering near the sliding surface can be avoided. The
system diagram is shown in Fig.2.2-2 and Fig.2.2-3.

www.intechopen.com
74 Electric Vehicles  Modelling and Simulations

Fig. 2.2-2. Schematic diagram of switch region of combined control

Contoller

VSC Strategy

Tm
System Status
E-motor Vehicle
Identification

MFC Strategy

ω,Vx,Tm

Fig. 2.2-3. Combined control block diagram

The simulation results for the vehicle that starts on the road surface with a low adhesion
coefficient =0.2 is shown in Fig.2.2-4.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 75

Fig. 2.2-4. Start on a low adhesion surface( =0.2)

From Fig.2.2-4, we can see that, on a low adhesion coefficient road surface, the vehicle
doesn’t slip. The slip ratio is in the ideal scope. Comparing with the above mentioned VSC
strategy, the fluctuation of the slip ratio for combined control is improved. The fluctuation
time continues 2.5s before stable convergence range occurs and the peak of the fluctuation of
the slip ratio is 0.5. With the work of the combined control strategy the fluctuation scope is
narrowed and the same to the e-motor’s output torque. The drive performance for the
combined control strategy is also excellent. On the low adhesion surface, the longitudinal
velocity can reach 17m/s after 10s from starting.
Table 1 shows the driving performance for different control methods on road surface with
low adhesion coefficient ( =0.2).

www.intechopen.com
76 Electric Vehicles  Modelling and Simulations

Utilization of
Drive Average
Time(s) adhesion
distance(m) acceleration(m/s2)
coefficient
Without Control 36 7.5 1.33 68%
MFC 29 5.8 1.72 88%
Combined control 26 5.7 1.75 89%
Table 1. Accelerate to 10m / s on road surface with low adhesion coefficient from starting
Fig. 2.2-5 displays the simulation results on the jump road from =0.2 to =0.7 . The
tire doesn’t slip on this kind of surface. It demonstrates that the combined control strategy is
effective to such surface, too.

Fig. 2.2-5. Start on the jump road from =0.2 to =0.7

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 77

Fig.2.2-6 shows the simulation results with MFC strategy which is on the low adhesion
coefficient road surface. In this simulation test, the wheel speed is disturbed that is manually
offset by white noise(0.1kw) in order to verify the effectiveness to the disturb of the velocity
signal.

Fig. 2.2-6. Effectiveness to the disturb of the wheel speed signal for MFC control

www.intechopen.com
78 Electric Vehicles  Modelling and Simulations

From Fig.2.2-6, we can get that MFC has not good performance to the disturb of the wheel
speed because the output drive torque is out of control. As we all known, the wheel speed is
the only control parameter to this kind of strategy so that MFC control isn’t regarded as an
excellent control method to realize ASR function.
Fig.2.2-7 shows the simulation results on the low adhesion coefficient road surface which
the longitudnal velocity is disturbed with combined control method. In this paper the
longitudinal velocity is manually offset by a positive 2m/s and white noise. According to
the results, the character of the combined control strategy is confirmed.

Fig. 2.2-7. Effectiveness to the disturb of the longitudinal velocity signal

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 79

According to results from Fig.2.2-6 and Fig.2.2-7, we can get that the combined control
method has better robustness to the input signal’s disturb. This point is very important to
the usage of the control method.

3. Anti-lock brake control

For electric vehicles, the motor inside each wheel is able to provide braking torque during
deceleration by working as a generator. Moreover, the torque response of an electric motor
is much faster than that of a hydraulic system. Thanks to the synergy of electric and
hydraulic brake system, the performance of the ABS (Anti-lock Brake System) on board is
considerably improved.
In this section, a new anti-skidding method based on the model following control method is
proposed. With the new feedback function and control parameter, the braking performance,
especially the phase-delay of the electric motor's torque is, according to the result of the
simulation, improved. Combined with the advantage of the origin MFC, the improved MFC
can be widely applied in anti-skidding brake control.
Furthermore, a braking torque dynamic distributor based on the adjustable hybrid braking
system is designed, so that the output torque can track the input torque accurately.
Meanwhile a sliding mode controller is constructed, which doesn’t perform with the slip
ratio value as the main control parameter. Accordingly, the total torque is regulated in order
to prevent the skidding of the wheel, so that the braking safety can be guaranteed.

3.1 Model following controller

When braking, slip ratio  is generally given by,

3.1.1 One wheel model

Vw  V

V
Where V is the vehicle longitudinal velocity and Vw is the wheel velocity. Vw=Rw, where R,
w are the wheel radius and angular velocity respectively.

Fig. 3.1-1. One wheel model dynamic analysis

www.intechopen.com
80 Electric Vehicles  Modelling and Simulations

In the light of Fig. 3.1-1, the motion equations of one wheel model can be represented as

  Fd  R  Tb
dw I w dVw
Iw (3.1-1)
dt R dt

 Fd
dV
Mw (3.1-2)
dt
In these equations, air resistance and rotating resistance are ignored. Mw is the weight of
one wheel; IW is the wheel rotational inertia; Tb is the braking torque, i.e. The sum of the
hydraulic braking torque and the braking torque offered by the electric motor, and Fd is the
braking force between the wheel and the road surface.

3.1.2 Design of MFC controller

The slip ratio is an important measurement for wheel's braking performance. For practical
vehicle, it is difficult to survey this velocity. Therefore the slip ratio is hard to obtain.
Compared with usual anti-skidding method, the method MFC(model following control) does
not depend on the information-slip ratio. Consequently it is beneficial for the practical use.
According to the result by Tokyo University:
For the situation-skidding, the transmit function is P(s )skid  w  
V 1 1
Fbrake Mw s

For the situation-adhesion, the transmit function is P(s )adh  

Vw 1 1
Fbrake M / 4  Mw s
The equation above is used as the nominal model in designing the controller “Model
Following Controller”. M represents the mass of the vehicle. Applying the controller, the
dynamics of the going to be locked wheel becomes close to that of the adhesive wheel,
through which the dynamics of the vehicle will be in the emergency situation.

3.1.3 Improved MFC controller

The above listed method, especially the feedback function is based on the one-wheel-model,
but in fact there is always load-transfer for each wheel so that it cannot appropriately reflect
the vehicle’s state. According to the origin feedback function for one-wheel-model
(M/4+Mw), which is introduced in the above-mentioned text, the information of the vertical
load of each wheel can be used to substitute for (M/4+Mw). Here it is called equivalent
mass and then the controller will automatically follow the state of the vehicle, especially for
acceleration and deceleration situation.
The specific way to achieve this idea is to use each wheel’s vertical load Fz to represent its
equivalent weight. So the feedback function should be Fz/g instead of (M/4+Mw).When
necessary, there should be a wave filter to obtain a better effect.
Another aspect ,which needs mo modify is its control parameter. For the method above, the
control parameter is the wheel velocity Vw. In order to have a better improvement of the
dw
braking performance, the wheel angular acceleration as the control parameter is taken
dt
advantage of.
R
It  M / 4 * R 2
Therefore the feedback function accordingly should be .

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 81

R
It  Fz / g * R 2
With the idea of the equivalent mass, the feedback function should be .

The reason why we take use of this control parameter is the electric motor itself also shows a
dw
delay (5~10ms) in an actual situation while the phase of the wheel angular acceleration
dt
precedes that of the wheel velocity Vw. Consequently this control method can compensate
the phases-delay of the electric motor.

3.1.4 Simulation and results

3.1.4.1 Simulation results with the wheel velocity as the control parameter
In the simulation, the peak road coefficient in the longitudinal direction is set to 0.2, which
represents the low adhesive road. The top output torque of the electric motor is 136Nm and
the delay time due to the physical characteristic of the electric motor 5 ms.
Fig. 3.1-2 shows the simulation result using the wheel velocity Vw as the control parameter.
The braking distance is apparently decreased. The slip ratio is restrained under 20%. The
unexpected increased amplitude of the slip ratio is mainly due to the delay of the electric
motor’s output, which can be proved in Fig. 3.1-2 (b). This can cause contradiction in the
braking process. Fig. 3.1-2 (c) shows longitudinal vehicle velocity and wheel velocity under
this control parameter.

(a) (b)

(c)
Fig. 3.1-2. Simulation Result of the Hybrid-ABS with the wheel velocity as the control
parameter

www.intechopen.com
82 Electric Vehicles  Modelling and Simulations

3.1.4.2 The simulation results with the angular acceleration as the control parameter
dw
Fig. 3.1-3 shows the simulation result using the wheel angular acceleration as the
dt
control parameter and increase the top output torque of the electric motor. Compared with
the previous simulation result, it is clear that the braking distance is further shortened
(compared with the system without electric motor control). The slip ratio is also restrained
under 20% and is controlled better that the previous control algorithm. From Fig. 3.1-3 (b)
we can see the phase-delay of the electric motor is greatly improved so that the two kinds of
the torques can be simply coordinated regulated.

(a) (b)

(c)
Fig. 3.1-3. Simulation results of the Hybrid-ABS with the angular acceleration as the control
parameter
Table 2 shows the result of the braking distance and the braking time under three above-
mentioned methods.

Hydraulic ABS without Hybrid ABS Hybrid ABS with

motor control with MFC improved MFC
Braking
27.9 26.8 26.5
distance(m)
Braking time(s) 5.12 4.87 4.83
Table 2. Results of the braking distance and the braking time under three different methods

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 83

3.1.5 Conclusion
According to the simulation results, the braking performance of the improved MFC is better
than the performance of the origin MFC, proposed by Tokyo University. In future can we
modify the MFC theory through the choice of the best slip ratio, because we know the value
of the best slip ratio is not 0 but about 2.0. When we can rectify MFC theory in this aspect,
the effect of the braking process will be better.

3.2 Design of the braking torque dynamic distributor

The distributor's basic design idea is to make the hydraulic system to take over the low
frequency band of the target braking torque, and the motor to take over the high frequency
band. Then the function of the rapid adjustment can be reached.

Fig. 3.2-1. The block diagram of the braking torque dynamic distributor
According to Fig. 3.2-1, C1(s) and C2(s) in Fig. 3.2-1 are the model of motor and hydraulic
system. They can be written expressed as (1) and (2):

C1 (s) 
 Ms  1
1
(3.2-1)

C 2 (s ) 
 Hs  1
1
(3.2-2)

Here,  M and  H are time constants for motor and hydraulic system relatively.
In order to reach the goal to track the braking torque, GSISO(s) =1, that is,

C 1 ( s)  G1 ( s )  C 2 (s )  G2 (s )  1 (3.2-3)

We can put formula (3.2-1) and formula (3.2-2) into formula (3.2-3),

C motor (s )   C hyd (s )  
 Ms  1  Hs  1 s  1
1 1 1
(3.2-4)

C motor (s )  [  C hyd (s )  ]  ( M s  1)
s  1  Hs  1
1 1

 s1  s1
(3.2-5)
 M  C hyd (s )  M
s  1  Hs  1

www.intechopen.com
84 Electric Vehicles  Modelling and Simulations

Here, τ is the sampling step

Chyd(s) is chosen as the second-order Butterworth filter, and then according to (3.2-5) we can
get Cmotor(s). And the saturation torque of the motor is limited by the speed itself.

3.3 Design of the sliding mode controller

3.3.1 Design of switching function
The control target is to drive the slip ratio to the desired slip ratio. Here a switching function
is defined as:

s    reference (3.3-1)

The switching function is the basis to change the structure of the model. And the commonest
way to change the structure is to use sign function- sgn(s). The control law here combines
equivalent control with switching control so that the controller can have excellent
robustness in face with the uncertainty and interference of the environment.
So the control law can be expressed as:

u  ueq  uvss (3.3-2)

Therefore the braking torque can be represented as:

Tb  Tb , eq  T sgn(s ) (3.3-3)

In practical engineering applications, the chattering may appear when sign function is used.
Therefore the Saturation function ‘sat ()’ is used to substitute for sign function.

Fig. 3.3-1. Saturation function

So the braking torque can be expressed as:

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 85

Tb  T b , eq  Tsat(
s

) (3.3-4)

3.3.2 The improved sliding mode controller

One desired slip ratio can’t achieve the best braking effect because of the inaccurate

based on sliding mode control will be proposed according to the characteristic of the   
measurement of the vehicle speed and the change of the road surface. Then, a new method

curve. It can seek the optimal slip ratio automatically. The typical    curve is shown in
Fig.3.3-2.

Fig. 3.3-2.    curve

d
From Fig. 3.3-2, we can see:
 0 ,   reference ,  needs increasing in order to obtain larger  . At this point we
d
When

d
can increase the braking torque on the wheel;
 0 ,   reference ,  needs maintaining in order to obtain larger  . At this point
d
When

d
we can maintain the braking torque on the wheel;
 0 ,   reference ,  needs decreasing in order to obtain larger  . At this point we
d
When

can decrease the braking torque on the wheel.

According to the one wheel model and the definition of slip ratio, we can receive:

 
d d / dt T I w V
   b w  x
d d / dt FZ R
Rw (3.3-5)
 
Tb  I w w
 
Vx

FZ R
w

www.intechopen.com
86 Electric Vehicles  Modelling and Simulations

That is:

 
Tb  I w w
When  
 0 ,  < reference , s    reference <0
w

 
Tb  I w w
When  
 0 ,  = reference , s    reference =0
w

 
Tb  I w w
When  
 0 ,  > reference , s    reference >0
w

 R  Vx
The interval of the optimal slip ratio is commonly from 0.1 to 0.2. Therefore, when the slip
ratio calculated by   is larger than 0.3, we can judge that the current slip ratio is
Vx

So the algorithm based on    curve can be improved as:

surely larger than the optimal slip ratio. The output of the sign function is 1.

 R  Vx
When the slip ratio calculated by   is bigger than 0.3, then we know that the
Vx
actual slip ratio must be bigger than the optimal slip ratio, then the output of the sign

 R  Vx
function is 1;
When the slip ratio calculated by   is smaller than 0.3,
Vx
i. If |w  | w ，

 J w  Tb
  0 sgn( s )  1
   w


  
 J w  Tb  0 sgn(s )  1
   w

ii.  | w
If|w
Sign function maintains the output of the last step, that is: sgn(s )t  sgn(s )t  1 .

3.3.3 Simulation and results

Fig. 3.3-3 shows the effect of the braking torque dynamic distributor. Since the existence of
the saturation torque of the motor, it can’t track the input torque when the input torque too
large. When the demand torque is not too large, the braking torque dynamic distributor
illustrates excellent capability.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 87

Fig. 3.3-3. The character of the braking torque dynamic distributor

www.intechopen.com
88 Electric Vehicles  Modelling and Simulations

Fig.3.3-4 - Fig.3.3-6 is the simulation results, which get from the improved sliding mode
controller, and the initial velocity of the vehicle is 80km/h, the saturation torque of the

i. When adhesion coefficient   0.9:

motor is 180Nm：

Fig. 3.3-4. Simulation results on the road with   0.9

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 89

ii. When adhesion coefficient   0.2:

Fig. 3.3-5. Simulation results on the road with µ = 0.2

www.intechopen.com
90 Electric Vehicles  Modelling and Simulations

iii. When adhesion coefficient changes in 1st second from 0.2 to 0.9:

Fig. 3.3-6. The road adhesion coefficient changes from   0.2 to   0.9 at the 1st second

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 91

From Fig.3.3-4 -Fig.3.3-6, we know that, although this method doesn’t regard slip rate as the
main control information, this sliding mode can track the optimal slip ratio automatically.
That means, both the longitudinal adhesion force and the lateral adhesion force can be made
use of fully. Even on the road, whose adhesion coefficient increases suddenly, the controller
can also find the optimal slip ratio.
During the braking process, the torque offered by the motor and hydraulic system doesn’t
oscillate distinctly. It indicates, the hybrid-braking system can achieve target braking torque
actually.
Table 3 shows the braking distance and braking time on the different road. From the datum
we know the braking safety can be guaranteed with this anti-skidding controller.

Number Adhesion coefficient Braking distance(m) Braking time(s)

a) 0.9 33.99 2.71
b) 0.2 136.6 11.62
c) 0.2-0.9 50.23 3.47
Table 3. Braking distance and braking time on the different road

3.3.4 Conclusion
The braking torque dynamic distributor, which combines the merits of the two actuators
motor and hydraulic system, can track the demanded torque promptly and effectively. The
sliding mode controller has two sorts. One is to track the desired slip ratio, which is set
manually and the effect of the controller good. However, the measurement of the vehicle
velocity and the identification of the road limit the promotion of the usage. The other kind
of controller can seek the optimal slip ratio automatically. Through the result of the
simulation, the effectiveness of this controller is proved. It can have a wider range of
application.

4. Vehicle stability control

Many researchers in the last decade have reported that direct yaw moment control is one of
the most effective methods of active chassis control, which could considerably enhance the
vehicle stability and controllability. The direct yaw moment control of a traditional ICE
(Internal Combustion Engine) vehicle is based on the individual control of wheel braking
force known as the differential braking. However, for EVs, the generation of desired yaw
moment for stabilizing the vehicle under critical driving conditions can be achieved by rapid
and precise traction/braking force control of each in-wheel-motor.
In this section, a hierarchical vehicle stability control strategy is introduced.
The high level of the control strategy is the vehicle motion control level. A dynamic control
system of a 4 in-wheel-motored electric vehicle which improves the controlling stability
under critical situation is presented. By providing the method of estimating the cornering
stiffness and combining the controller with optimal control allocation algorithm, which
takes account of the couple characteristic of the longitudinal/lateral force for tire under
critical situation, the vehicle stability control system is designed. The double lane change
simulation was carried out to verify the validity of the control method. Simulation result
shows the proposed control method could stabilize the vehicle posture well under critical
condition. Compared with the LQR with fixed cornering stiffness, the feedback from

www.intechopen.com
92 Electric Vehicles  Modelling and Simulations

identifying cornering stiffness to correct the parameters of the controller helps a lot in
improving the robustness of the stability control.
The low level of the control strategy is the control allocation level, in which the longitudinal
force’s distribution is the focal point. Through the analysis of the tire characteristics under
the combined longitudinal and lateral forces, an effectiveness matrix for the control
allocation considering the longitudinal force’s impact on the lateral force was proposed.
Based on Quadratic Programming method the longitudinal forces on each wheel are optimal
distributed. The simulation results indicate that the proposed method can enhance the
vehicle handling stability, meanwhile the control efficiency is improved as well.

4.1 Vehicle dynamic control structure

Studies have shown that hierarchical control of the dynamics control method has a clear,
modular control structure, as well as better control robustness, which is easy for real vehicle
applications of the control algorithms. This hierarchical control architecture is widely
adopted by general chassis’s integrated control.VDC(vehicle dynamic control) introduces
the hierarchical control structure, as shown in Fig. 4.1-1, the upper level is the vehicle
motion control and the bottom level is the control allocation for each actuator.
The motion controller which belongs to the first level in the stability algorithm, collects the
signals from the steering wheel’s angle and the accelerator pedal, and calculates the
generalized forces required by the stability control, including the longitudinal forces FxT
and yaw moment M zT . The longitudinal forces can be directly calculated according to the
accelerator pedal signals. The yaw moment can be got by following the reference model.

Fig. 4.1-1. Vehicle dynamic control structure

The control allocation is the second level of the vehicle controller. It is responsible to convert
the "generalized forces" to the sub-forces on each actuator according to certain distribution
rules and under some external constraint conditions (such as the maximum output of the
motor and the road adhesion coefficient, etc.). And then to realize the optimum distribution
of the each motor’s torque. For a 4WD electric vehicle driven by 4 in-wheel-motors, the sub-
force on each actuator is just the tire longitudinal force formed by the motor’s output torque.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 93

4.2 Vehicle motion controller

The yaw moment control is based on the MFC (model follow control) method. As reference
model, the DYC model could keep slip angle zero for stability. The gain scheduling control
algorithm can revise the parameters real-timely through the cornering stiffness
identification to improve the adaptability of the algorithm to the environment and the
change of the model parameters. The variable structure control (VSC) is applied to design
control algorithm, for considering the strong robust characteristic during uncertainty. With
proposed non-linear vehicle model, a precise gain value for switch function will be
calculated, in order to reduce chattering effect.

4.2.1 Vehicle model

4.2.1.1 Linear vehicle model
The simplified linear two freedom model make the side slip angle and the yaw rate as its
state variables. As the control input, the yaw moment M zT is gained from the longitudinal
force allocation by the motors according to the required moment, the function is:

d
m V(   )  Fyf  Fyr (4.2-1)
dt

d
Jz  M y  M zT (4.2-2)
dt

The description of the state space is:

X  AX  E  Bu (4.2-3)

Here： x  [   ]T , u  M zT

 2(C f  C r ) 2(C f l f  C r lr ) 
  1  
A 
mV 2
 2(C f l f  C r lr ) 2(C f l f  C r lr ) 
mV

  
2 2

 Jz J zV 

 2C f 
  0
E  ,B   1 
 2C f l f   
mV
(4.2-4)
   J z 
 Jz 

M y  Fyf l f  Fyr lr represents the yaw motion caused by the lateral force acting on each wheel,
Fyf ， Fyr are the total front/rear wheel lateral forces. Other parameters are shown in
Fig.4.2-1.

www.intechopen.com
94 Electric Vehicles  Modelling and Simulations

Fig. 4.2-1. Planar vehicle motion model

4.2.1.2 Non-linear vehicle model
In this paper the arc-tangent function is used to fit the lateral force formula, then a simple
non-linear vehicle model can be obtained, the arc-tangent function contains two fitting
parameters c1 , c 2 , the fitting effect is show below:

as: x  [ x1 x2 ]T  [   ]T , h( x )    x1 ,  is the centroid-side angle of the vehicle,  is the

The state space of One-track non-linear vehicle model can be express

course angle of the vehicle, u means additional yaw moment input M zT , the complete
function is:

7000

6000

5000
tire lateral force/N

4000 c1*atan(c2*alfa)

3000

Magic model
2000
arc tangent function

1000

0
0 0.1 0.2 0.3 0.4 0.5 0.6
slip angle /rad

Fig. 4.2-2. Arc-tangent function vs. magic model

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 95

 1
 {c f 1 a tan[c f 2 ( f  x1  x2 )]cos  f
lf
 mv

V
 cr 1 a tan[cr 2 ( x1  x2 )]}  x2
 x 1  
lr

 x  
V
 2   {l c a tan[c (  x  l f x )]cos 
(4.2-5)

 Jz f f 1
1


f2 f 1 2 f
V
  lr cr 1 a tan[cr 2 ( x1  r x2 )]  u}

l
V
Here, m represents the mass of the vehicle, J z represents the yaw rotational inertia of the
vehicle, c f 1 and c f 2 are the fitting parameters for the front wheel, cr 1 and cr 2 are the fitting
parameters for the rear wheel, l f is the distance from the gravity point to the front axle and

vehicle,  f is the steering angle for the front wheel.

lr is the distance from the gravity point to the rear axle, V is the gravity point velocity of the

Based on non-linear model mention above, we can design yaw-rate follow controller. In our
case, the dynamic function of yaw rate is second-order system:

  f ( X , t )  f ( X , t )  g( u)  d(t )

 (Fyf ( f , Ffz ,  )  l f  Fyr ( r , Frz ,  )  lr )

1
(4.2-6)
Jz

 f ( X , t )  u  d(t )
1
Jz

Here,  f is the side slip angle for the front wheel,  r is the side slip angle for the rear

vertical load for the front and rear wheel,  is the road adhesion coefficient.
wheel, Fyf and Fyr are the side slip force for the front and rear wheel, Ffz and Frz are the

f ( X , t ) indicates non-linear system function; g(u) indicates non-linear continued function;

f ( X , t ) and d(t ) stand for uncertainty and external disturbance of controlled object, which
are supposed to be zeros.

4.2.2 Reference model

The desired yaw-rate output is calculated from the reference model (DYC):

d   d  
d d
1 kd
(4.2-7)

Jz V
Here: kd  ；d 
2C f V
mV  2(C f l f  C r lr )
2
2(C f l f 2  C r lr 2 )

4.2.3 Controller design

4.2.3.1 Gain scheduling controller
Based on the linear vehicle model, the controller adapts the LQR stability control algorithm.
It is composed of feed-forward and feedback. Supposing the relationship between the feed-
forward yaw moment and the front-wheel steering angle as:

www.intechopen.com
96 Electric Vehicles  Modelling and Simulations

M ff (s )  G ff  (s ) (4.2-8)

Here: G ff is the feed-forward yaw moment coefficient. It can be calculated through the

condition, i.e.  (s )  (s ) when  (0)  0 . Then.

transfer function from vehicle side slip angle to front-wheel steering angle under stable

4C f C r l f lr  2C f l f mV 2
G ff 
mV 2  2(C f l f  C r lr )
(4.2-9)

Feedback control is used to decrease the control system’s error caused by the unknown
perturbation and the imprecise of the model, and to improve the reliability of the control
system.
Define the state error E  X  X d , from function (4.2-3), (4.2-7):

E  AE  BM fb  ( A  Ad )X d  (E  Ed ) (4.2-10)

Considering the last two as perturbation, and according to LQR, assure the target function
below to be least:

J   (EQET  uRuT )dt


(4.2-11)
0

By solving the Riccati function ， feedback coefficient G fb is gained. And the feedback
moment is:

M fb  G fbE   g fb 1 (    d )  g fb 2 (   d ) (4.2-12)

Total yaw moment required is:

M zT  M ff  M fb (4.2-13)

From the analysis above, we know the total yaw moment is decided by the feed-forward
coefficient G ff and feed-back coefficient G fb together. And the coefficients can be adjusted
on time according to the front and rear cornering stiffness identified and the vehicle speed
measured. The control algorithm refers to the linear optimization calculation and on-line
resolution of the Riccati function, which can affect the real time performance. On the real car
the coefficients corresponding to different cornering stiffness and the vehicle speed are
calculated off-line previously. Then a look-up table will be made from that and will be
downloaded to the ECU for control. To easily show the movement of the feed-forward and
feed-back coefficients, the following figure will illustrate the change of the front and rear
cornering stiffness together through supposing the front cornering stiffness is changing,
while the rear one is a fixed proportion to it.
Cornering stiffness is an important parameter for the controller. It will change along with
the road condition or under the critical condition of the vehicle, which will further affect the
control precise of the vehicle stability. The cornering stiffness that DYC control relies on is
linear to the cornering stiffness under the current condition. So the cornering stiffness in this
paper is estimated based on the two freedom linear model.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 97

6
x 10
8

feed-forward gain 4

-2 0
0 5
10
20 4
10 x 10
30
vehicle velocity[m/s] 40 15 front tire cornering
stiffness[N/rad]

4
x 10
4
yaw rate feedback gain

0
0 40
5
20
4 10
x 10 vehicle velocity[m/s]
15 0
front tire cornering stiffness[N/rad]

0
beta feedback gain

-200

-400

-600

-800
0

5
4
x 10 40
10 30
front tire cornering 20
stiffness[N/rad] 15 10
0 vehicle velocity[m/s]

Fig. 4.2-3. Feed-forward/Feed-back Map

www.intechopen.com
98 Electric Vehicles  Modelling and Simulations

From function (4.2-2) , M y is:

M y  2C f (     )l f  2C r (  
lf lr
)lr (4.2-14)
V V
Here C f , C r are front and rear nominal cornering stiffness. M y above needs to be estimated
by the yaw moment observation(YMO) below:

ˆ  F(s )( J   M )
M (4.2-15)
y z zT

Here: F(s )  c /(s  c ) is a filter function to gain  . c is truncation frequency.

estimator must provide the information of  . Therefore a united estimation of C f , C r , 

From function (4.2-14): to estimate the front and rear cornering stiffness separately, the

needs to be established. To simplify the design, some change has been made to the function

load under a certain load range (C f l f  C r lr ) . And as  is a small value, then:

above. According to the magic tire model, the wheel cornering stiffness is pro rata to the

C f l f   C r lr   M y ； (4.2-16)

Thus function (4.2-14) can be ：

l f  lr
M y  2C f l f (   ) (4.2-17)
V

ˆ (t )   T   (t ), T  C ,
M (4.2-18)
y f

 l f  lr 
 (t )  F(s)   2 l f (    ) 
 
(4.2-19)
V

Based on the above model, the front and rear cornering stiffness C f , C r will be gained by
RLS estimation, as follows:

( k  1) ( k )
ˆ( k )  ˆ( k  1) 
   T ( k )( k  1) ( k )

  T ( k )ˆ( k  1)  y( k ) 
( k  1)  
1 
( k )   ( k  1) ( k ) ( k )( k  1) 

   T ( k )( k  1) ( k ) 
T (4.2-20)

 is forget factor and can be properly selected according to the road condition.
With the estimation result the controller parameters can be corrected on time. And a more
precise general force can be gained to improve the allocation control of the vehicle.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 99

4.3 Control alloction alogrithm

Through the control of the upper level, the yaw moment M zT is gained, which will be
allocated to each actuator to realize the control target (on 4WD EV is the motor torque).

Making approximation: sin   0 and cos   1 , the total vehicle longitudinal force and the
4.3.1 Effectiveness matrix

yaw moment caused by the longitudinal force are as follows:

FxT  Fxfl  Fxfr  Fxrl  Fxrr

M zxT  ( Fxfl  Fxfr  Fxrl  Fxrr )

b (4.3-1)
2

Expressed as:

FxT  Bx
F
M zT  Bzx F
x
(4.3-2)
x

 [Fxfl Fxrr ]T ; Bx  [1 1 1 1] , Bzx  [  

F

b b b b
Where: x Fxfr Fxrl ] , Bx and
2 2 2 2
Bzx are named as the effectiveness matrix.
In most researches, the vehicle yaw moment was directly obtained by (4.3-1). As the
coupling characteristics of tires, the change of the tire longitudinal forces leads to the change
of its’ lateral force, especially in the critical conditions. So it’s necessary to consider the

Under certain tire sideslip angle  , the relationship between the four wheels’ lateral and
additional yaw moment caused by the change of the lateral force.

longitudinal forces can be expressed as:

 f yx
F


y ( x) (4.3-3)

Where: y  [Fyfl Fyfr Fyrl Fyrr ]T

F


f yx is a non-linear function, which brings complexity in the computation of the effectiveness
matrix and the optimization of the control distribution. While if direct linear approximation
was made to it, it would be too simplistic.
Discretization of the total yaw moment demand from the vehicle motion controller comes
to:

M zT  t  1   M zT  t   M zT (4.3-4)

Supposing that  is a small value, then sin   0 and cos   1 . The increment of the total
yaw moment can be expressed as:

M zT  B zx Fx  B zy Fy (4.3-5)

 b b
Here: Bzx     ， Bzy  l f lr lr 
2 
b b
 2
lf
2 2

www.intechopen.com
100 Electric Vehicles  Modelling and Simulations

Fx   Fxfl Fxfr Fxrl Fxrr 

T

Fy   Fyfl Fyfr Fyrl Fyrr 

T

Under a certain tire cornering angle  , the coupling relation of the tire longitudinal/lateral
forces can be expressed as:

Fy  f yx

 Fx  (4.3-6)

Here: Fx  Fxfl Fxrr  , Fy  Fyfl Fyrr  .

T T
Fxfr Fxrl Fyfr Fyrl
then: Fy  f yx

 Fx 
Magic formula can describe the tire characteristics under the combined working condition,
but too complex. According to tire friction ellipse, the tire characteristics can be
 Fy   F 2
approximated expressed as:       x   1 , where Fy0 is lateral tire force under tire
2

 Fy 0   Fx max 
 
sideslip angle  when longitudinal force equal zero, and Fxmax is maximum longitudinal tire
force under tire sideslip angle  .

 Fyi 0 i 2  Fxi
 f  

 i j
  Fyi  Fxmax i 2

(4.3-7)
i j
yx

0
ij

To substitute function (4.3-5) with (4.3-7), then:

M z  ( Bzx  Bzy f yx

)Fx (4.3-8)

Set virtual control vector v   FxT M zT  ,

Then:
T

where the total longitudinal forces FxT are created by the driver’s pedal command. And the
actual control vector u  [  Fxfl Fxfr Fxrl Fxrr ]T . Then the control allocation should
meet the following equation:

v  Bu (4.3-9)

 
Where: the effectiveness matrix B    
Bx

 Bzx  Bzy f yx 

4.3.2 Optimal allocation algorithm

One objective of the control allocation can be expressed as to minimize the allocation error:

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 101

min Wv ( Bu  v )

s.t. u   u  u 
(4.3-10)

Wv is the weight matrix, reflecting the priority of each generalized force. The constraints
include the limited capacity of the actuator, ie. the maximum torque range of in-wheel-
motors, and the road adhesion ability.
At the same time, we also hope to minimize the energy consumption of the actuator.
Considering the characteristics of the tire adhesion, different wheels with different vertical
load Fz , then the longitudinal forces and the lateral forces provided by each wheel are not
the same. So the weight matrix Wu is introduced. It is a diagonal matrix, and the diagonal
elements are:

wii 
1
( ii Fzii )2  (Fxii  Fyii
(4.3-11)
2 2
)

Where  is the road adhesion coefficient of each wheel. Fx , Fy and Fz are the longitudinal
force, the lateral force and the vertical load of each wheel of the time.
Then another objective can be expressed as:

min Wu (u  ud )
s.t. u  u  u
(4.3-12)

Wu considerate the characteristic of each tire adhesion, because different wheel is with
different vertical load Fz .
The above (4.3-10) and (4.3-12) can be combined as followed Quadratic Programming (QP)
problem:

u  arg min  ( Wu (u  ud ) 2   W ( Bu   ) 2 )
2 2
(4.3-13)
u uu

Thus the computation time can be reduced largely. The parameter  is usually set to very
large in order to minimize the allocation error. The optimization problem can be solved
through active set methods.

4.4 Simulation results and analysis

Using vehicle dynamics analysis software veDYNA, combined with the proposed vehicle
stability control algorithm above, the high velocity double lane change operation is
simulated to verify the validity of the control algorithm.
The vehicle is to carry out double lane change operation with the velocity of about 100km/h,
which should be as constant as possible during the operation. Fig.4.4-2 shows the contrast
between the vehicle trajectories with and without stability control. The vehicle could keep a
steady posture and avoid obviously lateral slippage. Meanwhile compared to the LQR control
without identification of the cornering stiffness, the algorithm designed in this paper can
decrease the impact of the change of the model’s parameters on the control effect. In addition a
little under steering during lane change presents the steering characteristic of DYC reference
model to restrain over large side slip angle.

www.intechopen.com
102 Electric Vehicles  Modelling and Simulations

Fig. 4.4-1. veDYNA Simulation Model

Double lane change

LQR control
5

4
without control
y-position [m]

1
0 with estimation LQR control

-1

-2
0 20 40 60 80 100
x-position [m]
Fig. 4.4-2. Vehicle Trajectory

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 103

0.6

0.4

gradient of slip-angle(rad/s)
1

0.2
0.5
yaw -rate(rad/s)

0
-0.2

-0.5 -0.4
actual yawrate
desired yawrate
-1
8 10 12 14 16 18 -0.2 -0.1 0 0.1 0.2
time(s) slip-angle(rad)

10 6

4
Lateral acceleration [m/s2]

Roll angle [deg] 2

0
0

-5
-2

-10 -4
0 1 2 3 4 5 0 1 2 3 4 5
Time [s] Time [s]

Fig. 4.4-3. Vehicle States

Fig. 4.4-3 presents the behaviors of several state values of the vehicle during such operation.
Among them the yaw rate response can match the desired value well. Supposing on level
and smooth road, when the peak value of the lateral acceleration is close to 1.0g, the vehicle
has been working under critical condition. The    phase trajectory indicates that the


vehicle can keep steady even when the slip angle reaches 8 degree.

4
x 10
3
cornering stiffness[N/rad]

2.5

1.5

front tire estimated value

0.5
rear tire estimated value
0
5 10 15 20
time[s]

Fig. 4.4-4. Estimated Cornering Stiffness of Tire

www.intechopen.com
104 Electric Vehicles  Modelling and Simulations

Fig. 4.4-5. Actual vs Desired Yaw Moment

Fig. 4.4-4 shows the estimated values of the cornering stiffness in the double lane change
simulation, the vehicle lateral motion characteristic adjusts acutely during lane change. If
the LQR controller were designed according to the fixed value of the cornering stiffness, the
control effect would get worse along with the fluctuation of the cornering stiffness. Fig. 4.4-5
illustrates how the actual yaw moment follows the requirement of the control during the
whole control process. It’s clear that the optimize allocation algorithm can finely meet the
requirement of the stability control even under the critical condition.

0.5
switch function value

-0.5

-1
10 11 12 13 14 15 16 17
time[s]

Fig. 4.4-6. Switch function value

Figure shows the adjustment of switch function value during operation; there is obvious
chattering when simulation time is near 12s and 14s, but that causes no severe fluctuate to
general yaw moment. Figure illustrates the actual yaw moment can realize the general
control requirement basically, which guarantees the achievement of motion follow control.

www.intechopen.com
Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle 105

4.5 Conclusion
In this section, a motion follow controller is designed based on the linear optimal control
theory with the real time identification of the cornering stiffness and VSC theory with non-
linear vehicle model. Considering the longitudinal/lateral forces’ couple characteristic of
tire under critical condition, the optimal control allocation algorithm realized the vehicle
stability control successfully. Using veDYNA software, the double lane change simulation is
carried out to verify the validity of the control method above.
The results of the simulation show: the proposed control method could stabilize the vehicle
posture well under critical condition (i.e. the peak lateral acceleration is close to 1.0g and the
slip angle reaches 8 degree). The reference model plays an important role on restraining the
vehicle from losing stability. Compared with the LQR with fixed cornering stiffness, the
feedback from identifying cornering stiffness to correct the parameters of the controller
helps a lot in improving the robustness of the stability control. Actual control allocation
could match the general requirement well by taking into account of tire force characteristic.

5. References
[1] D. Foito, A. Roque, J. Esteves, J. Maia, Electric Vehicle with two Independent Wheel
Drives-Improving the Performance with A Traction Control System, Proceedings
on the “17th International Electric Vehicle Symposium & Exposition - EVS 17”,
Montreal, Canada, October 2000.
[2] Buckholtz K R. Reference Input Wheel Slip Tracking Using Sliding Mode Control [R].
SAE 2002 World Congress, 2002.
[3] Zhaoliang Zhang, Lu Xiong and Zhuoping Yu. The Anti-skidding Control for
Combined- braking System based on theadjustable Hydraulic System. 2010 IEEE
International Conference on VehicularElectronics and Safety, Qingdao, 2010.
[4] Yoichi HORI ， Future Vehicle driven by Electricity and Control -Research on Four
Wheel Motored “UOT Electric March II”, IEEE, Vol 51, 2004 P954 – 962.
[5] Shin-ichiro Sakai，Takahiro Okano, Tai Chien Hwa， 4 Wheel Motored Vehicle ”UOT
Electric March II” -Experimental EV for Novel Motion Control Studies-, Proc. of
INTERMAC2001 Joint Technical Conference, Tokyo, Japan, 2001.
[6] Yoichi Hori, Yasushi Tyoda, and Yoshimasa Tsuruoka. Traction Control of Electric
Vehicle: Basic Experimental Results Using the Test EV “UOT Electric March”. IEEE
Transactions on Industry Applications, vol.34, pp.1131-1137, September/October
1998.
[7] Buckholtz K R. Reference Input Wheel Slip Tracking Using Sliding Mode Control [R].
SAE 2002 World Congress, 2002.
[8] Y.Chamaillard, G.L. Gissinger, J.M.Perronne and M.Renner. An original braking
controller with torque sensor. Proceedings of the Third IEEE Conference on Control
Applications, 1994.
[9] A.El Hadri, J. C. Cadiou, K. N.M’Sirdi and Y. Delanne. Wheel-slip regulation based on
sliding mode approach. SAE 2001 World Congress, 2001.UC Berkeley, November
1993.
[10] Kachroo P. Nonlinear Control Strategies and Vehicle Traction Control [D].
Ph.D .dissertation

www.intechopen.com
106 Electric Vehicles  Modelling and Simulations

[11] Shino,M.,Wang,Y.,Nagai,M., Motion Control of Electric Vehicle Considering Vehicle

Stability. Proc.of AVEC2000,2000, pp.705-711
[12] Motoki Shino, Masao Nagai, "Yaw moment control of electric vehicle for improving
handling and stability", JSAE Review 2001, pp.473-480
[13] Peng He and Yoichi Hori, “Optimum Traction Force Distribution for Stability
Improvement of 4WD EV in Critical Driving Condition”, 9thIEEE International
Advanced Motion Control, Workshop, Istanbul, 2006
[14] Zhuoping Yu, Wei Jiang and Lijun Zhang, "Torque distribution control for four wheel
in-wheel-motor electric vehicle", Tongji Daxue Xuebao/Journal of Tongji
University, Vol. 36, No. 8, 2008, pp.1115-1119
[15] Sakai,S.,Hori,Y., Advanced Vehicle Motion Control of Electric Vehicle Based on The
Fast Motor Torque Response. Proc.of AVEC2000, 2000 ,pp .729-736
[16] Chumsamutr,R., Fujioka,T., Improvement of Electric Vehicle’s Cornering Performance
by Direct Yaw Moment Control. Proc. of AVEC2000, 2000, pp.729-736
[17] H .Fujimoto, N .Takahashi, A. Tsumasaka and T. Noguchi. motion control of electric
vehicle based on cornering stiffness estimation with yaw-moment observer. IEEE
Int. Workshop Advanced Motion Control, 2006,pp. 206-211
[18] Peng He and Yoichi Hori, “Optimum Traction Force Distribution for Stability
Improvement of 4WD EV in Critical Driving Condition”, 9th IEEE International
Advanced Motion Control, Workshop, Istanbul, 2006
[19] Zhuoping Yu, Wei Jiang and Lijun Zhang, "Torque distribution control for four wheel
in-wheel-motor electric vehicle", Tongji Daxue Xuebao/Journal of Tongji
University, Vol. 36, No. 8, 2008, pp.1115-1119
[20] H.B.Pacejka, "Tyre and Vehicle Dynamics", Butterworth Heinemann, 2002
[21] Brad Schofield and Tore Haegglund, "Optimal Control Allocation in Vehicle Dynamics
Control for Rollover Mitigation", 2008 American Control Conferance, Washington,
2008
[22] O. haerkegard, "Backstepping and control allocation with applications to flight control",
Ph.D. dissertation, Linkping University, 2003

www.intechopen.com
 Electric Vehicles - Modelling and Simulations
Edited by Dr. Seref Soylu

ISBN 978-953-307-477-1
Hard cover, 466 pages
Publisher InTech
Published online 12, September, 2011
Published in print edition September, 2011

In this book, modeling and simulation of electric vehicles and their components have been emphasized
chapter by chapter with valuable contribution of many researchers who work on both technical and regulatory
sides of the field. Mathematical models for electrical vehicles and their components were introduced and
merged together to make this book a guide for industry, academia and policy makers.

How to reference
In order to correctly reference this scholarly work, feel free to copy and paste the following:

Lu Xiong and Zhuoping Yu (2011). Vehicle Dynamic Control of 4 In-Wheel-Motor Drived Electric Vehicle,
Electric Vehicles - Modelling and Simulations, Dr. Seref Soylu (Ed.), ISBN: 978-953-307-477-1, InTech,
Available from: http://www.intechopen.com/books/electric-vehicles-modelling-and-simulations/vehicle-dynamic-
control-of-4-in-wheel-motor-drived-electric-vehicle

InTech Europe InTech China

University Campus STeP Ri Unit 405, Office Block, Hotel Equatorial Shanghai
Slavka Krautzeka 83/A No.65, Yan An Road (West), Shanghai, 200040, China
51000 Rijeka, Croatia
Phone: +385 (51) 770 447 Phone: +86-21-62489820
Fax: +385 (51) 686 166 Fax: +86-21-62489821
www.intechopen.com