Sie sind auf Seite 1von 6

See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/257644645

A Mini-Baja Active Suspension Modelling using


ADAMS/Car
CONFERENCE PAPER MAY 2005

4 AUTHORS, INCLUDING:
Ricardo A. Ramrez-Mendoza

Ruben Morales-Menendez

Tecnolgico de Monterrey

Tecnolgico de Monterrey

219 PUBLICATIONS 199 CITATIONS

200 PUBLICATIONS 276 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Ricardo A. Ramrez-Mendoza


Retrieved on: 30 August 2015

A MINI-BAJA ACTIVE SUSPENSION MODELLING USING ADAMS/CAR


Aline Driveta, Ricardo A Ramrez-Mendozaa, Rubn Morales-Menndezb and Fernando Peaa.
a
ITESM Monterrey Campus, Mechatronics Engineering Program
b
Center for Innovation in Design and Technology
Ave. Garza Sada Sur # 2501 Monterrey, Nuevo Len, Mxico
{ricardo.ramirez, drivet,rmm} @itesm.mx

KEY WORDS
Skyhook-control, modelling and simulation, Mini-Baja
vehicles, active suspension and vehicle dynamics .
ABSTRACT
This paper presents a study of the modelling of a Minibaja vehicle. The full-scale nonlinear model of the vehicle
was performed with a commercial software. Additionally,
some active suspension control laws have been tested to
emphasize the performance of the proposed model.

1. Introduction
Active suspension has been studied over the past few
years; several control strategies have been proposed and
simulated with numeric modelling software. Nowadays,
advanced active suspension systems are acquiring more
importance due to the need of improving manoeuvrability
and comfort in a vehicle. The role of a suspension is to
isolate the chassis and the passengers from unevenness of
the road, that is, to avoid vibration, without decreasing the
manoeuvrability of the car. There are several articles in
the literature where ASS (active suspension system) is
applied [1]. However, all research has been conducted in
a quarter of a vehicle, and using simplified models of the
car. In the present paper, the aim is to model the full
vehicle with powerful three-dimensional software witch
can model the complete dynamic of the vehicle. When a
more complex model can be used or a more detailed
vehicle then, more accurate results can be obtained from
applying a simple control strategy is applied [2]. Two
control strategies are used, the first one is the skyhook
control [3], and the classical proportional + integral +
derivative controller approach also known as PID
controller. This research however, was limited, to the
modelling of Mini-baja vehicle in ADAMS/Car [4]
where the engine was considered as ideal, the body of the
vehicle as rigid and no deflection was considered. The
model used is based on a model provided by the MSC
ADAMS/Car database [5]. The model used was updated
and modified to meet the requirements of the test.

2. Skyhook Control

A suspension system is used to support the vehicle


weight, to isolate the chassis from road disturbances. The
goal is to improve the quality of the ride and to maintain
the traction force between the tire and the road [6]. There
are two kinds of suspension: active suspension and
passive suspension.
A passive suspension system is one in which the
characteristics of the components (springs and dampers)
are fixed. The designer of the suspension establishes the
vehicles characteristics where a compromise between the
stability and the comfort of the vehicle is established.
An active suspension on the other hand will adjust to the
road type improving both comfort and stability. To obtain
an active suspension it is necessary to vary the damper
characteristics along with the road profile or use an
actuator, which can vary a force according to a control
law.
One of the most recurrent strategies in the literature is the
skyhook control, which adjusts the damping level of the
suspension. The most common is to control by changing
the damping constant between two values using the so
called on-off control since the damper switches back and
forth between two possible damping states. When the
chassis is moving up, and the tires are getting farther , the
damping constant should be low if not zero, when chassis
is moving down and the tires are getting closer, the
damping constant should be large (ideally an infinite
damping constant) The effect of the skyhook control
scheme is to minimize the absolute vertical velocity of the
chassis. The skyhook strategy is a simple approach that
has been proven to work for a quarter of a vehicle [7].
The skyhook strategy requires a varying force. This force
can be obtained using a hydraulic damper varying the size
of the orifice in the hydraulic flow valve. Using an
electro-rheological damper or a magneto-rheological
damper and then applying various levels of electric or
magnetic field will result in a change of the viscosity of
the fluid. While changing the viscosity of the fluid it
adjusts the required damper force [8], [9].
Karnopp [10] proposed a model in which a damper is
placed in the suspension to suppress the vibratory motion
of the sprung mass, and indeed, it reduced the resonant
peak of the sprung mass but was unable to control the
resonant peak of the un-sprung mass. A modified skyhook
proposed by J. Karl Hedrick [11] which shall be
considered in order to minimize both resonant peaks. In
this study the vertical velocity of the chassis is measured

and then used to compute the varying force that is applied


to the chassis by the actuator. Suspension can be modified
in the four wheels of the vehicle, in this study however
only the two front wheels are chosen, equation (1) is used
to compute the force applied:

F (t ) Vz (t ) K

(1)

Where F (t ) is the force applied by the actuator,


the vertical velocity of the chassis, and
gain for process.

Vz (t ) is

K is the adequate

The skyhook strategy requires a variable damper constant


which could be calculated using an equation using a
velocity feedback and multiplying it with a constant.
Now, to apply such control it is necessary to use a damper
that can change its constant over a certain range of
values. In Figure # 1 is shown the configuration of the
active suspension:

Fig # 1 . Model of an active suspension with a skyhook


strategy.

3. Mini-Baja Solid Model


Most of the studies realized in active suspension have
been modelled in a quart of a vehicle in Matlab software,
a more accurate study can be conducted with
ADAMS/Car software applying a skyhook control
strategy [12].
Using ADAMS/Car to analyse a solid or threedimensional prototype is similar to test a physical
prototype. To run a test in ADAMS/Car it is necessary
to specify most of the components, or subsystems, that
form a full car, to mention some of them, it is necessary to
fully determine the suspension, the steering and braking
system, the body or chassis, materials and tires [13]. The
model used is a model of the database of MSC
ADAMS/CAR that has been modified and adapted,
particularly the dimensions of certain elements, the
materials, weight of the suspension, steering and frame.

ADAMS/Car contains full-vehicle test rigs and several


tests can be applied to the full-vehicle, but also any
measure can be user defined as, for example the velocity
in any point of the body of the vehicle. ADAMS/ Car is
a dynamic simulator, which can fully evaluate the
behaviour of a specific model in different environments.
Due to the extent of detail necessary to use ADAMS/
Car the Mini -Baja vehicle was chosen. It offers the
advantage to be a simple car, easier to model with
ADAMS/CAR with greater accuracy and fully detailed.
Also the Mini-Baja model used in this research is based
on a physical model that the ITESM uses in competitions
of the SAE (Society of Automotive Engineers). The SAE
competition is held to promote research and development
of automotive engineering. The following picture shows
the real live car:

Fig # 2. Mini-Baja real vehicle used to make the


ADAMS model.
The more basic systems were redefined and adjust in
order to have a very precise model of the real car, just to
mention a few: brake system, steering system, powertrain
and so on.
In the used model, the vehicles front suspension system
is controlled through varying the force of the actuator in
order to vary its reaction force. This force will be
determined through a skyhook control algorithm. The
suspension used is a wishbone type where there are: a
spring a damper and the force that will simulate the
variable damper, all the members are solid (no deflection
is considered). The full suspension is displayed in Figure
# 3.

Fig # 3 Modelling of the Mini-Baja suspension system.


Once all the subsystems are specified, the full assembly
can be determined obtaining the model shown in Figure
#4

Fig.# 5. Markers used to determine measures in tests.


The test applied to the model was a rundown through an
uneven road at a 40-km/hr constant speed. The road was
computer generated to be irregular and match the
measured stochastic profiles that are typical. Both sides
have no correlation.

The displacement obtained with a skyhook strategy is in


average lower and the resonant peaks of the chassis are
eliminated as we can see in Figure # 6

Fig # 4. Modelling of the Mini-Baja full vehicle


assembly.

4. Tests and simulations results


To prove that the active suspension improves the
performance of the vehicle, tests were conducted with the
ADAMS/Car software. These tests were applied to a
model of a vehicle with passive suspension and reapply to
the same vehicle with a modified suspension and having a
skyhook control strategy. During the test, velocity
acceleration and force on the spring and damper were
measured to determine the best performance. In order to
measure the velocity, acceleration and force, it is
necessary to specify a specific point in the model where
the variables will be determined. These are called markers
in ADAMS/CAR. Several markers were necessary to
measure velocity, displacement and acceleration as it is
shown in Figure # 5

Fig #6. Vertical Displacement of the chassis


Evidently, the more the vehicle has movement in the
vertical axe the less the passenger is comfortable,
particularly, he will resent the resonant peaks, which are
eliminated. Another important measure is the
acceleration, which is important to determine the quality
of the road. It was also observed an overall improvement
of an average 30% improvement and the resonant peaks
were also eliminated, which severely deteriorate the
comfort a manoeuvrability of the vehicle. The
improvement of the acceleration can be observed in the
Figure#8.

Fig.7. Vertical Acceleration of the chassis.

5. Using a Modified PID Controller


Fig # 9 Vertical displacement of the chassis
The PID is a widely used control in the continuous
industrial processes. In this process however instead of
using the error as a feedback and variable to determine the
force, the variable used is the vertical velocity. The
equation of the PID applied to the vertical velocity is
equation (2)

F (t ) K (v(t )
Where

There is also improvement in the vertical velocities of the


chassis, the skyhook has a 35% less velocity in the chassis
improving the quality of the ride, but the PID has 50%
improvement as we can see in Figure #9.

1
dv
v(t ) Td ) (2)

Ti
dt

K , Ti , Td are the PID parameters for the process.

The tuning was obtained by trial and error in simulation


and the results reported were the best trial.
The results are conclusive, the PID improves the
performance of the vehicle, the vertical acceleration is
diminished in more than 50% as it can be appreciated in
the next chart:

Fig.10.Vertical velocity of the chassis.

5. Conclusion

Fig # 8. Vertical Acceleration of the chassis


Displacement is also improved; the vehicle is almost
steady as we can see in Figure # 9

The most important purpose of choosing an active


suspension system is to improve the ride quality and the
manoeuvrability of the vehicle. To improve the ride
quality, it is important to isolate the sprung mass from the
unevenness of the road and to eliminate the vertical
vibrations, especially the resonant peaks which are
particularly uncomfortable to the passenger. The PID has
proven to be a better strategy than skyhook in the overall
performance. The force needed to obtain such results is
not greater than any of the suspension components (as the
spring or damper), which can be easily obtained. There
are also important results in flat roads where the PID had
a better performance. Some simplifications were made in
the model of the vehicle and to improve the accuracy of
the results a more precise model can be develop, however
the results are relevant for a comparison between two
control strategies.

References:
[1] Hrovat, D., and Hubbard, M., "Optimum Vehicle
Suspensions Minimizing RMS Rattlespace, Sprung-mass
Acceleration, and Jerk," Journal of Dynamic Systems,
Measurement, and Control, Vol. 103, No. 3, pp. 228-236,
September 1981.
[2] Yuchun Zhang; Lin Zhao; Hua Cong; Baohua Wang,
Study on control of vehicle attitude and ride comfort
based on full-car model, WCICA 2004. Fifth World
Congress on Intelligent Control and Automation 2004,
Volume: 4, 15-19 June 2004. Pages: 3514 - 3519.
[3] Hong, K. S., Sohn, H. C., and Hedrick, J. K., Modified
Skyhook Control of Semi-Active Suspensions: A New
Model, Gain Scheduling, and Hardware-in-the-Loop
Tuning, ASME Transactions, Journal of Dynamic
Systems, Measurement, and Control, Vol. 124, No. 1, pp.
158-167, March 2002.
[4]
Software
form
MSC.
ADAMS/CAR.
www.mscsoftware.com.au/products/software/msc/adams/
[5] Database intended for ADAMS/Car Users.
http://www.mscsoftware.com/support/university/student_
competitions/templates/templates_main.cfm.
[6] Margolis, D.L., "A Procedure for Comparing Passive,
Active, and Semiactive Approaches to Vibration
Isolation," Journal of the Franklin Institute, Vol. 315, No.
4, pp. 225-238, April 1983.
[7] Sergio M. Savaresi., Enrico Silani, Sergio Bittanti,
Semi-active suspensions: an optimal control strategy for a
quarter-car model; IFAC Symposium on ADVANCES IN
AUTOMOTIVE CONTROL, University of Salerno, Italy April 19-23, 2004.
[8] Lazareva, T.G. and Shitik, I.G., "Magnetic and
Magnetorheological Properties of Flowable Compositions
Based on Barium and Strontium Ferrites and Iron Oxides,
"Proceedings of the Society for Optical Engineering, Vol.
3040, pp. 185-189, March 1997.
[9] Kordonsky, W., "Elements and Devices Based on
Magnetorheological Effect,"Journal of Intelligent
Materials, Systems, and Structures, Vol. 4, pp. 65-69,
January 1996.
[10] Karnopp, D. C., Crosby, M. J., and Harwood, R. A.,
Vibration Control Using Semi-Active Force Generators,
ASME J. Eng. Ind., 96, No. 2, pp. 619626. 1974
[11] Mark D. Donahue, J. Karl Hedrick, Theses:
Implementation of an Active Suspension, Preview
Controller for Improved Ride Comfort, The University of
California at Berkeley, April 2001, Pages 6:26.

[12] Octavio Cesar Rangel Gil, Theses: Modelacin,


identificacin y control de suspensiones automotrices de
diferente estructura cinemtica. ITESM, 2002.
[13] Kruczek, A.; Stribrsky, A., A full-car model for
active suspension - some practical aspects. Mechatronics,
2004. ICM '04. Proceedings of the IEEE International
Conference on, June 3-5, 2004. Pages: 41 45