MULTIBODY DYNAMICS: BRIDGING

FOR MULTIDISCIPLINARY
APPLICATIONS
Jorge A.C. Ambr´ osio ´
IDMEC Instituto Superior Tecnico, Av. Rovisco Pais 1, 1049-001 Lisboa, Portugal ´
Abstract Simple or complex systems characterized by large relative motions be-
tween their components find in the multibody dynamics formalisms the
most general and efficient computational tools for their analysis. Ini-
tially restricted to the treatment of rigid bodies, the multibody methods
are now widely used to describe the system components deformations,
regardless of their linear or nonlinear nature. The ease of including in
the multibody models different descriptions of the contact problems,
control paradigms or equations of equilibrium of other disciplines is
demonstrated here to show the suitability of these approaches to be
used in multidisciplinary applications
Keywords: Flexible multibody dynamics, contact, biomechanics, vehicle dynamics,
railway dynamics, crashworthiness.
1. Introduction
The design requirements of advanced mechanical and structural sys-
tems and the real-time simulation of complex systems exploit the ease
of use of the powerful computational resources available today to cre-
ate virtual prototyping environments. These advanced simulation facili-
ties play a fundamental role in the study of systems that undergo large
rigid body motion while their components experience material or geo-
metric nonlinear deformations, such as vehicles, deployable structures,
space satellites, machines operating at high speeds or robot manipula-
tors. Some examples of engineering and biological systems for which
the large overall motion is of fundamental importance are exemplified
in Fig. 1. If on the one hand the nonlinear finite element method is the
most powerful and versatile procedure to describe the flexibility of the
system components, on the other hand the multibody dynamic formula-
tions are the basis for the most efficient computational techniques that
61
W. Gutkowski and T.A. Kowalewski (eds.), Mechanics of the 21st Century, 61–88.
© 2005 Springer. Printed in the Netherlands.
62 ICTAM04
Figure 1. Natural biological and artificial engineering systems for which multibody
dynamics provides irreplaceable modeling methodologies.
deal with large overall motion. Therefore, it is no surprise that many
of the most recent formulations on flexible multibody dynamics and on
finite element methods with large rotations share some common features.
In multibody dynamics methods, the body-fixed coordinate frames are
generally adopted to position each one of the system components and
to allow for the specification of the kinematic constraints that represent
the restrictions on the relative motion between the bodies. Several for-
malisms are published suggesting the use of different sets of coordinates,
such as Cartesian [1], natural [2] and relative coordinates [3]. Depend-
ing on the type of applications, each of these types of coordinates has
advantages and disadvantages. Due to the ease of the computational
implementation, their physical meaning and the widespread knowledge
of their features, all the formalisms presented in this work are based on
the use of Cartesian coordinates.
The methodological structure of the equations of motion of the multi-
body system obtained allows the incorporation of the equilibrium equa-
tions of a large number of disciplines and their solution in a combined
form. The description of the structural deformations exhibited by the
system components by using linear [5] or non-linear finite elements [6] in
the framework of multibody dynamics is an example of the integration
of the equations of equilibrium of different specialties. Of particular
importance for the applications pursued with the methodologies pro-
posed is the treatment of contact and impact, which is introduced in
the multibody systems equations by using either unilateral constraints
Multibody Dynamics: Bridging for Multidisciplinary Applications 63
[7] or a continuous contact force model [8]. The availability of the state
variables in the multibody formulation allows for the use of different
control paradigms in the framework of vehicle dynamics, biomechanics
or robotics and their integration with the multibody equations [9]. The
coupling between the fluid and structural dynamics equations allows for
the development of applications, where the fluid-structure interaction is
analyzed, especially for cases with large absolute or relative rotations in
the system components, are of importance [10, 11].
The research carried at IDMEC provides the examples offered in this
work. Application cases involving the modeling of realistic mechanisms,
passive safety of road and rail vehicles, impact and human locomotion
biomechanics, automotive and railway dynamics are used to demonstrate
the developments reviewed here.
2. Rigid Multibody Dynamics
A multibody system is defined as a collection of rigid and/or flexible
bodies constrained by kinematic joints and eventually acted upon by
a set of internal and/or external forces. The position and orientation of
each body i in the space is described by a position vector r
i
and a set of
rotational coordinates p
i
, which are organized in a vector as [1]:
q
i
= [r
T
, p
T
]
T
i
. (1)
According to this definition, a multibody system with nb bodies is
described by a set of coordinates in the form:
q = [q
T
1
, q
T
2
, . . . , q
T
nb
]
T
. (2)
The dependencies among system coordinates, which result from the
existence of mechanical joints interconnecting several bodies, are defined
through the introduction of kinematic relationships written as [1]:
Φ(q, t) = 0, (3)
where t is the time variable, which is used only for the driving con-
straints. The second time-derivative of Eq. (3) with respect to time
yields:
¨
Φ(q, ˙ q, ¨ q, t) = 0 ≡ D¨ q = γ, (4)
where D is the Jacobian matrix of the constraints, ¨ q is the acceleration
vector and γ is the vector that depends on the velocities and time.
The system kinematic constraints are added to the equations of mo-
tion using the Lagrange multipliers technique [1]. Denoting by λ the
64 ICTAM04
vector of the unknown Lagrange multipliers, the equations of motion for
a mechanical system are written as
_
M D
T
D 0
_ _
¨ q
λ
_
=
_
f
γ
_
(5)
where M is the global mass matrix, containing the mass and moments
of inertia of all bodies, and f is the force vector, containing all forces
and moments applied to the system bodies plus the gyroscopic forces.
The Lagrange multipliers, associated with the kinematic constraints,
are physically related with the reaction forces generated between the
bodies interconnected by kinematic joints, given by [1]
f
(c)
= −Φ
T
q
λ, (6)
The usual procedures to handle the integration of sets of differential-
algebraic equations must still be applied in this case in order to eliminate
the constraint drift or to maintain it under control [1, 2].
Forward Dynamics
The computational strategy used to solve the forward dynamics of
the system, represented by Eq. (5), is outlined in Fig. 2. The solution
procedure starts by the determination of the initial positions and velo-
cities of the system components. Next, the system inertia, the Jacobian
matrices, the forces and the right-hand-side of the kinematic accelera-
tion constraint equations vectors, are calculated and assembled in the
equations of motion. Equation (5) is then solved to find the system ac-
celerations, and in the process the Lagrange multipliers. By integrating
the current velocities and the system accelerations, at time t, the new
positions and velocities for time t +∆t are calculated by using a variable
Figure 2. Solution of the forward dynamic analysis of a multibody system.
Multibody Dynamics: Bridging for Multidisciplinary Applications 65
order, variable time-step integration procedure [1, 2, 4]. The forward dy-
namics simulation proceeds until the previously set final time is reached.
The procedure outlined in Fig. 2 is used in general purpose multibody
dynamics codes, such as DAP-3D [1]. Throughout this work it is demon-
strated that all engineering applications foreseen here are implemented,
either by developing specific kinematic constraints or by implementing
force models in Eq. (5).
Application Example of a Roller Coaster. When a body travels
along a guide, not only its path has to be followed, but also its spa-
tial orientation has to be prescribed, according to spatial characteristics
of the curve. The formulation adopted to implement these kinematic
constraints, using the moving Frenet frame associated with the track
centerline based on the work by Pombo and Ambrosio [12], is outlined ´
next.
Prescribed Motion Constraint. The objective here is to define the
constraint equations that enforce that a point of a rigid body follows the
reference path [12]. Consider a point R, located on a rigid body i, that
has to follow the specified path depicted in Fig. 3. The path is defined
by a parametric curve g(L), which is controlled by a global parameter
L that represents the length travelled along the curve until the current
location of point R. The kinematic constraint is
Φ
(pmc,3)
= 0 ≡ r
R
i
−g(L) = 0, (7)
where r
R
i
= r
i
+A
i
s

R
i
represents the coordinates of point R with respect
to the global coordinate system (x, y, z), r
i
is the vector that defines
the location of the body-fixed coordinate system (ξ, η, ζ)
i
, A
i
is the
transformation matrix from the body i fixed coordinates to the global
reference frame, and s

R
i
represents the coordinates of point R with re-
Figure 3. Local frame alignment constraint.
xxxxxxx
66 ICTAM04
spect to the body-fixed reference frame. For notational purposes (·)

means that (·) is expressed in body-fixed coordinates.
The second part of the constraint ensures that the spatial orientation
of body i remains unchanged with respect to the moving frame of the ref-
erence path represented in Fig. 3. Consider that (u
ξ
, u
η
, u
ζ
)
i
represent
the unit vectors associated with the axis of the body-fixed coordinate
system (ξ, η, ζ)
i
. Let the Frenet frame of the general parametric curve
g(L) be defined by the principal unit vectors (t, n, b)
L
. At the ini-
tial time of analysis, the relative orientation between the body vectors
(u
ξ
, u
η
, u
ζ
)
i
and the local frame (t, n, b)
L
leads to
Φ
(lfac,3)
= 0 ≡


⎧⎧

⎨⎨n
T
· u
ξ
b
T
· u
ξ
n
T
· u
ζ


⎫⎫

⎬⎬



⎧⎧

⎨⎨a
b
c


⎫⎫

⎬⎬
= 0. (8)
This kinematic constraint ensures that the alignment remains constant
throughout the analysis. The transformation matrix from the body i
fixed coordinates to the global coordinate system is:
A
i
= [u
ξ
u
η
u
ζ
]
i
(9)
defining the following unit vectors as:
u
1
= {1 0 0}
T
; u
2
= {0 1 0}
T
; u
3
= {0 0 1}
T
. (10)
Equation (8) is now rewritten in a more usable form as:
Φ
(lfac,3)
= 0 ≡


⎧⎧

⎨⎨n
T
A
i
u
1
b
T
A
i
u
1
n
T
A
i
u
3


⎫⎫

⎬⎬



⎧⎧

⎨⎨a
b
c


⎫⎫

⎬⎬
= 0, (11)
which constitutes the second part of the path following constraint.
Roller-Coaster Dynamics. Let the roaller-coaster rail be defined
with the spatial geometry described in Fig. 4. The path-following con-
straint is used to enforce the vehicles to follow the rail for the prescribed
geometry.
The roller coaster vehicle consists of a train with three cars that are
interconnected by linking bars, represented in Fig. 5. The multibody
model of the vehicle is assembled using eleven rigid bodies, corresponding
to 3 car bodies, 6 wheelsets and 2 connection bars.
The complete vehicle model only has 1 d.o.f., which is the longitudinal
motion of the cars. The motion of the vehicle is guided by the dynamics
described by Eq. (5). A view of the motion of the roller coaster is dis-
played in Fig. 6 and the details of the analysis are found in reference [13].
Multibody Dynamics: Bridging for Multidisciplinary Applications 67
Figure 4. View of the roller coaster as used in the simulation.
Figure 5. Multibody model of the roller coaster vehicle.
Figure 6. Snap shots of the roller coaster motion as observed from the second car.
Note that the study of these vehicles only requires the use of the path-
following constraint. The contact forces are not explicitly used but they
can be calculated from the Lagrange multipliers associated to the path
constraint.
Inverse Dynamics
In many applications all external forces are known and the motion of
the system is also known. Therefore, the only unknowns are the internal
forces. Let the first row of Eq. (5) be re-written as
M¨ q +D
T
λ = g (12)
z
3
1 2 4 5
7
x
9
6
8
10
11
68 ICTAM04
which is
D
T
λ = g −M¨ q. (13)
Equation (13) emphasizes that the only unknowns of the system are
the Lagrange multipliers. The reaction forces at the joints are given by:
g
(c)
= −D
T
λ. (14)
The solution of the equations of motion in inverse dynamics can be
used to solve for the internal forces of the human body, i.e., muscle and
anatomical joint reaction forces, that develop for known motions.
Application to Biomechanics: Gait Analysis
For biomechanical applications in gait a three-dimensional model, pre-
sented in Fig. 7, is used [14]. It has a kinematic structure made of thirty-
three rigid bodies, interconnected by revolute and universal joints, in
such a way that sixteen anatomical segments are represented.
3
11 1
16 16
5
12 2
2
13
7
44
6
9
1
10
15
14
8
2
3
4
5
6
7
8
9
10
11
12
1
13
14
15
16
v1
v2
v3
v3
v7
vv88
v6
v6
v5
v5
v4
vv4
v211
v99
v10
v11
v11
vvv22
v12
v12
v13
v13
vv15
v14
v15
v16
v16
v17
v20 20
v20
v17
v18
vv19 99
14
15
16
17
18
19
20
21
22
23
7
6
8
4
5
2
3
1
9
10
11
12
13
31
32 25
24 26
27
28
29
30
33
Figure 7. The biomechanical model, its kinematic structure and a detail of the
ankle joint.
Joint Moments-of-Force: A Determinate Problem. To drive
the biomechanical model in the inverse dynamic analysis, joint actuators
such as the one represented in Fig. 8 for the knee joint, are specified. The
actuators are the kinematic constraints in which the angle between two
adjacent segments is a known function of time. These additional equa-
tions are added to the system kinematic equation so that the number of
Multibody Dynamics: Bridging for Multidisciplinary Applications 69
m
1
m
2
m
3
O
m1
I
m1
I = O
m2
I
m2
=
2
O
m3
I
m3
Figure 8. Joint actuator associated with the knee joint and muscle actuator.
non-redundant constraint equations becomes equal to the number of co-
ordinates. Equation (13) is solved to obtain the Lagrange multipliers as-
sociated with the joint actuators, representing the net moments-of-force
of the muscles that cross those joints. The inverse dynamics problem,
as stated here, is totally determined.
Muscle Forces: A Redundant Problem. The solution of the
inverse dynamics problem with muscle actuators introduces indetermi-
nacy in the biomechanical system, since it involves more unknowns than
equations of motion. By using optimization techniques to find the mus-
cle forces that minimize a prescribed objective function, a solution for
the problem is obtained. The optimization problem is stated as:
minimize F
0
(u
i
)
subject to


⎧⎧

⎪⎪

⎨⎨

⎪⎪
f
j
f (u
i
) = 0, j = 1, ..., n
ec
,
f
j
f (u
i
) 0, j = (n
ec
+ 1) , . . . , n
tc
,
u
lower
i
u
i
u
upper
i
i = 1, . . . , n
sv
(15)
where u
i
are the state variables bounded respectively by u
lower
and
u
upper
, F
0
(u
i
) is the objective or cost function to minimize and f
i
ff (u
i
)
are constraint equations that restrain the state variables.
The minimization of the cost functions simulate the physiological cri-
teria adopted by the central nervous system when deciding which muscles
to recruit and what level of activation to obtain the adequate motion.
Several cost functions have been proposed for the study of the redundant
problem in biomechanics [15]. The minimization of the sum of the cube
of the muscle stresses [16] is often used in applications involving human
70 ICTAM04
Figure 9. Lower extremity muscle apparatus.
Figure 10. Muscle forces for the hamstrings and triceps surae.
Multibody Dynamics: Bridging for Multidisciplinary Applications 71
locomotion
F
0
=
n
ma

m=1
_
¯ σ
F
m
l
FF F
m
˙
l
FF
F
m2
0
FF
a
m
_
3
(16)
where n
ma
are the number of muscle actuators and ¯ σ is the specific
muscle strength with a constant value of 31.39 N/cm
2
[17]. The human
locomotion apparatus, represented in Fig. 9, is modeled having the mus-
cles with the physiological data described in Yamaguchi [17]. The state
variables associated with muscle actuators represent muscle activations
that can only assume values between 0 and 1.
To illustrate the type of results obtained for the muscle forces in a case
of normal cadence gait of a 50%ile male, the muscle forces for the ham-
strings and triceps surae are presented in Fig. 10.
3. Contact and Impact
Let a triangular patch, where point k of the body shown in Fig. 11
will impact, be defined by points i, j and l. The normal to the outside
surface of the contact patch is defined as n = r
ij
×r
jl
. The position of
the point k with respect to point i of the surface is
r
ik
= r
k
−r
i
(17)
which is decomposed in a tangential and a normal component, given by
r
t
ik
= r
ik

_
r
T
ik
n
_
n; r
n
ik
=
_
r
T
ik
n
_
n. (18)
The necessary conditions for contact are that node k penetrates the
‘front’ surface of the patch, but not through its ‘back’ surface, with
which a thickness e is associated. These conditions are written as
0 r
T
ik
n e. (19)
n
jl
r
j
li
r
l
ij
r
ij
ik
r
i
( ) ( ) ( ) n r
ik
nnn
ik
rr =
( ) n r r
ik ik
rr
tt
ik
rr − =
i
j
l
k
k
*
Figure 11. Contact detection between a finite element node and a surface.
72 ICTAM04
The remaining necessary conditions for contact result from the need
for the node to be inside of the triangular patch. These three extra
conditions are
_
˜r
t
ik
r
ij
_
T
n 0;
_
˜r
t
ik
r
jl
_
T
n 0 and
_
˜r
t
ik
r
ki
_
T
n 0. (20)
Equations (19) and (20) are necessary conditions for contact. How-
ever, depending on the contact force model actually used, they may not
be sufficient to ensure effective contact.
Unilateral Constraints
If contact between a node and a surface is detected, a kinematic con-
straint is imposed. For flexible bodies let us assume a fully plastic nodal
contact, i.e., the normal components of the node k velocity and acceler-
ation, with respect to the surface, are null during contact
˙ q
k
= ˙ q
(−)
k

_
˙ q
(−)T
k
n
_
n; ¨ q
k
= ¨ q
(−)
k

_
¨ q
(−)T
k
n
_
n (21)
where ˙ q
(−)
k
and ¨ q
(−)
k
represent the nodal velocity and acceleration imme-
diately before impact respectively . The kinematic constraint implied by
Eqs. (21) is removed when the normal reaction force between the node
and the surface becomes opposite to the surface normal, i.e.,
f
n
k
f = −f
T
k
f n > 0. (22)
It should be noted that the contact force is related to the Lagrange
multiplier associated by the kinematic constraint defined by Eqs. (22).
Therefore, the change of sign of the force is in fact the change of sign of
the multiplier.
This contact model is not suitable to be used directly with rigid body
contact. The sudden change of the rigid body velocity and acceleration
would imply that the velocity and acceleration equations resulting from
the kinematic constraints would not be fulfilled. Other forms of this
contact model can be found in the work by Pfeiffer and Glocker [7].
Continuous Contact Force Model
An alternative description of contact considers this to be a continuous
event where the contact force is a function of the penetration between
the surfaces. This leads to the continuous force contact model, proposed
by Lankarani and Nikravesh [8], and briefly described here. Let the
contact force between two bodies be written as
f
s,i
f =
_

n
+ D
˙
δ
_
u (23)
Multibody Dynamics: Bridging for Multidisciplinary Applications 73
where δ is the pseudo-penetration,
˙
δ is the pseudo-velocity of penetra-
tion, K is the equivalent stiffness, D is a damping coefficient and u is
a unit vector normal to the impacting surfaces. Using the hysteresis dis-
sipation model and the equivalent stiffness, calculated for instance for
Hertzian elastic contact [18], the nonlinear contact force is
f
s,i
f = Kδ
n
_
1 +
3
_
1 −e
2
_
4
˙
δ
˙
δ
(−)
_
u (24)
where
˙
δ
(−)
is the initial contact velocity and e is the restitution coeffi-
cient. Note that K is a function of the geometry and material properties
of the impacting surfaces.
Application to Railway Dynamics – The Wheel-Rail
Contact Problem
One of the interesting applications of multibody dynamics with con-
tact mechanics is the description of the wheel-rail contact in railway
dynamics, represented in Fig. 12. The stability of the running vehicle
depends ultimately on the rail-wheel contact and on the vehicle primary
suspension. Therefore, methodologies that provide accurate models to
represent the phenomena are of particular importance.
In a general case of a railway vehicle one or two points of each wheel
are in contact with the rail, as shown in Fig. 12. The diametric sec-
tion that contains the wheel flange contact point makes an angle s
f
Rw
with the diametric section that contains the wheel tread contact point.
The possibility of detecting contact in different diametric sections allows
predicting derailment and it is, therefore, of utmost importance.
Let the generalized geometry of the rail and wheel be described by
generalized surfaces resulting from sweeping the rail profile along the rail
Figure 12. Two points of contact in the rail and wheel surfaces: lead contact.
Tr Tr
Fla Fla Fla Fla
(Le (Le (Le (Le
Tr Tr Tr Treeee
con con con con
ct ct ct ct
ct) ct) ct) ct)
74 ICTAM04
centerlines and the wheel profiles around the base circle of the wheel.
In order to ensure that the search for the contact points is between
convex surfaces, the wheel profile is divided in treat and flange profiles.
The contact between the rail and one of the wheel surfaces is described
generically in Fig. 13, where the mating surfaces are represented as free
surfaces.
Figure 13. Candidates to contact points between two parametric surfaces.
The geometric conditions for contact between the convex surfaces are
defined by vector products defined between the surfaces. The first condi-
tion is that the surfaces normals n
i
and n
j
at the candidates to contact
points have to be parallel. This condition means that n
j
has null pro-
jections over the tangent vectors t
u
i
and t
w
i
:
n
j
×n
i
= 0 ⇔


⎧⎧

⎨⎨
n
T
j
t
u
i
= 0,
n
T
j
t
w
i
= 0.
(25)
The second condition is that the vector d, which represents the dis-
tance between the candidates to contact points, has to be parallel to the
normal vector n
i
. This condition is mathematically written as:
d ×n
i
= 0 ⇔


⎧⎧

⎨⎨
d
T
t
u
i
= 0,
d
T
t
w
i
= 0.
(26)
The geometric conditions (25) and (26) provide four nonlinear equations
with four unknowns, the four parameters u, w, s and t that define the
two surfaces. This system of equations provides solutions for the location
of the candidates to contact points that have to be sorted out.
x
Multibody Dynamics: Bridging for Multidisciplinary Applications 75
The coordinates of the candidates to contact points are determined
by solving an optimal problem and the distance between such points is
calculated in the process. The points are in contact if
d
T
n
j
0. (27)
When contact is detected, the normal force is calculated using Eq. (24)
and the tangential forces are evaluated using the Kalker theory, the
Polach formulation or the Heuristic nonlinear creep model. It has been
found that the Polach formulation provides the best approach for the
tangent forces, and it is used hereafter [13].
The wheel-rail contact model outlined here is used to model the ML95
trainset, shown in Fig. 14, which is used by the Lisbon subway company
(ML) for passengers’ traffic.
The multibody model of the trailer vehicle of the train, developed
in the work by Pombo [13], is composed of the car shell suspended by
a set of springs, dampers and other rigid connecting elements on the
bogies. This assembly of connective elements constitutes the secondary
suspension, sketched in Fig. 15, which is the main one responsible for
the passenger’s comfort.
The connections between the bogies chassis and the wheelsets, also
achieved by another set of springs, dampers and rigid connecting ele-
Figure 14. Schematic representation of the ML95 trainset.
Figure 15. Secondary suspension model of the ML95 trailer vehicle.
76 ICTAM04
Figure 16. Primary suspension model of the ML95 trailer bogie: a) Three-
dimensional spring-damper elements; b) Suspension model with springs and dampers.
Figure 17. Lift of the right wheel of the leading wheelset for vehicle forward velo-
cities of 10 and 20 m/s, using the Kalker linear theory.
ments, constitute the primary suspension represented in Fig. 16. The
primary suspension is the main suspension responsible for the vehicle
running stability.
The simulation results of the vehicle, running in a circular track with
a radius of 200 m with velocities of 10 and 20 m/s, show that the predic-
tion of flange contact is of fundamental importance. Fig. 17 shows that
contact forces obtained with the Kalker linear theory originate the lift of
the outer wheel of the front wheelset at the entrance of the curve. De-
spite this wheel lift, derailment does not occur and the analysis proceeds
up to end. Nevertheless, such results are not realistic since the existence
of flange contact involves high creepages, which makes the Kalker linear
theory inappropriate to compute the creep forces. Therefore only the
Heuristic and the Polach creep force models must be considered.
Another aspect to note is that flange contact is detected with all creep
force models. Even when running at the speed of 10 m/s, where the cen-
trifugal forces effect is balanced by the track cant, flange contact occurs.
Lateral flange forces develop on the wheels of both wheelsets of the front
bogie as presented in Fig. 18 for a vehicle forward velocity of 10 m/s and
a)
Three-dimensional
spring-damper elements
Bogie frame Axlebox
Wheelset
b)
Multibody Dynamics: Bridging for Multidisciplinary Applications 77
using the Polach creep force model. During curve negotiation, the outer
wheel of the leading wheelset and the inner wheel of the rear wheelset
have permanent flange contact.
Referring to Fig. 19, for the velocity of 10 m/s, the flange contact oc-
curs on the outer and in the inner wheels of the vehicle. For the velocity
of 20 m/s, only the outer wheels have flange contact. This is explained
by the fact that, when running at 20 m/s, the vehicle negotiates the
curve with a velocity higher than the balanced speed.
4. Flexible Multibody Dynamics with Plastic
Hinges
Many applications of multibody dynamics require the description of
the flexibility of its components. For structural crashworthiness it is
-5 000
0
5 000
10 000
15 000
20 000
25 000
0 3 6 9 12 15 18 21 24 27 30
Time [s]
L
a
t
e
r
a
l

F
l
a
n
g
e

F
o
r
c
e

[
N
]
Left Ws 3 ( Pol ach)
Right Ws 3 ( Pol ach)
Left Ws 4 ( Pol ach)
Right Ws 4 ( Pol ach)
Figure 18. Lateral flange forces on the wheels of both wheelsets of the front bogie
for a vehicle forward velocity of 10 m/s, using the Polach creep force model.
Front
wheelset
(Ws 4)
Flange
contact
Rear
wheelset
(Ws 3)
Flange
contact
Figure 19. Contact configuration during curve negotiation.
78 ICTAM04
often unfeasible to use large nonlinear finite element models. The use
of multibody dynamics with plastic hinges is an alternative formulation
that allows building insightful models for crashworthiness.
Formulation of Plastic Hinges
In many impact situations, the individual structural members are
overloaded giving rise to plastic deformations in highly localized regions,
called plastic hinges. These deformations, presented in Fig. 20 for stru-
ctural bending, develop at points where maximum bending moments
occur, load application points, joints or locally weak areas [19]. Multi-
body models obtained with this method are relatively simple, which
makes the procedure adequate for the early phases of vehicle design.
The methodology described herein is known in industry as conceptual
modeling [20].
Figure 20. Localized deformations on a beam and a plastic hinge.
The plastic hinge concept has been developed by using generalized
spring elements to represent constitutive characteristics of localized plas-
tic deformation of beams and kinematic joints to control the deforma-
tion kinematics [21], as illustrated in Fig. 21. The characteristics of
the spring-damper that describes the properties of the plastic hinge are
obtained by experimental component testing, finite element nonlinear
analysis or simplified analytical methods.
The plastic hinge constitutive equation can be modified to account
for the strain rate sensitivity of some materials. A dynamic correction
factor is used to account for the strain rate sensitivity given by [21].
P
d
PP /P
s
PP = 1 + 0.07V
0.82
0
VV , (28)
where P
d
PP and P
s
PP are the dynamic and static forces, respectively, and V
0
VV is
the relative velocity between the adjacent bodies. The force or moment
Multibody Dynamics: Bridging for Multidisciplinary Applications 79
Analytical
Test
Rotation ( Rad)
M
o
m
e
n
t

(
k
N
m
)
0 .05 .10 .15 .20 .25
0
1
5
3
0
4
5
6
0
Figure 21. Plastic hinge bending moment and its constitutive relationship.
to apply due to the plastic hinge is multiplied by the ratio calculated in
Eq. (28) before it is used in the force vector of the multibody equations
of motion.
Application of the Plastic Hinge Approach to
Crashworthiness of Surface Vehicles
The multibody of an off-road vehicle with three occupants, shown in
Fig. 22, is used to demonstrate the plastic hinge approach to complex
crash events. The model includes all moving components of the vehicle,
suspension systems and wheels, and a tire model [16]. The biomechanical
models for the occupants are similar to those described in Fig. 7.
The three occupants, with a 50%tile, integrated in the vehicle are
seated. Two occupants in the front of the vehicle have shoulder and lap
Figure 22. Initial position of the vehicle and occupants for the rollover.
80 ICTAM04
seat belts while the occupant seated in the back of the vehicle has no
seatbelt.
The rollover situation for the simulation is such that the initial condi-
tions correspond to experimental conditions where the vehicle moves on
a cart with a lateral velocity of 13.41 m/s until the impact with a water-
filled decelerator system occurs. The vehicle is then ejected with a roll
angle of 23 degrees.
The results of the simulation are pictured in Fig. 23 by several frames
of the animation. The vehicle first impacts the ground with its left tires.
At this point the rear occupant is ejected. The rollover motion of the
vehicle proceeds with an increasing angular velocity, mainly due to the
ground – tire contact friction forces. The occupants in the front of the
vehicle are held in place by the seat belts. Upon continuing its roll
motion, the vehicle impacts the ground with its rollbar cage, while the
ejection of the rear occupant is complete. Bouncing from the inverted
position, the vehicle completes another half turn and impacts the ground
with the tires again. The HICs for all occupants largely exceed 1000,
which indicates a very high probability of fatal injuries for the occupants
under the conditions simulated.
An experimental test of the vehicle was carried out at the Transporta-
tion Research Center of Ohio [22], being an overview of the footage ob-
tained shown in Fig. 24. The outcome of the experimental test, which
is rather similar to the outcome of the simulation, is further used to
validate the vehicle model [21].
Figure 23. Views of the outcome of the rollover simulation of a vehicle with three
occupants.
Figure 24. View of the experimental test for the truck rollover.
Multibody Dynamics: Bridging for Multidisciplinary Applications 81
5. Flexible Multibody Dynamics with Finite
Elements
General Equations of Motion of a Single Body
Let the principle of the virtual works be used to express the equi-
librium of the flexible body in the current configuration t+∆t and let
an updated Lagrangean formulation be used to obtain the equations of
motion of the flexible body [23]. Let the finite element method be used
to represent the equations of motion of the flexible body. Referring to
Fig. 25, the assembly of all finite elements used in the discretization of
a single flexible body results in its equations of motion written as [6]


M
rr
M
rf
M
rf
M
φr
M
φφ
M
φf
M
fr
M

M
ff




¨r
˙ ω

¨ u



=


⎡⎡
g
r
g

φ
g

f





⎡⎡
s
r
s

φ
s

f





⎡⎡
0
0
f






0 0 0
0 0 0
0 0 K
L
+K
NL


⎤⎡


0
0
u




(29)
where ¨r and ˙ ω

are respectively the translational and angular accelera-
tions of the body-fixed reference frame and ¨ u

denotes the nodal accele-
rations measured in body fixed coordinates. The local coordinate frame
ξηζ attached to the flexible body, is used to represent the gross motion
of the body and its deformation.
Figure 25. General motion of a flexible body.
ted updated
ation
′ t
b
∆ b
t
b
t t
h
t
ζ
t
ηη
tt ∆
ζ
t ∆
00
ζζ
0
η
82 ICTAM04
Linear Deformations of Flexible Bodies
In many situations it is enough to consider that the components of
the multibody system experience only linear elastic deformations. Fur-
thermore, assume that the mode superposition technique can be used.
Then, the flexible part of the body is described by a sum of selected
modes of vibration as
u

= Xw (30)
where the vector w represents the contributions of the vibration modes
towards the nodal displacements and X is the modal matrix. Due to the
reference conditions, the modes of vibration used here are constrained
modes. Due to the assumption of linear elastic deformations the modal
matrix is invariant. The reduced equations of motion for a linear flexible
body are [5]
_
M
r
M
rf
X
X
T
M
fr
I
__
¨ q
r
¨ w
_
=
_
g
r
X
T
g
f
_

_
s
r
X
T
s
f
_

_
0
Λw
_
(31)
where Λ is a diagonal matrix with the squares of the natural frequencies
associated with the modes of vibration selected. For a more detailed
discussion on the selection of the modes used the interested reader is
referred to [5].
The methodology is demonstrated through the application to the si-
mulation of the unfolding of a satellite antenna, the Synthetic Aperture
Radar (SAR) antenna that is a part of the European research satellite
Figure 26. The European satellite with the folded and unfolded configurations of
the antenna.
Multibody Dynamics: Bridging for Multidisciplinary Applications 83
ERS-1, represented in Fig. 26. During the transportation the antenna is
folded, in order to occupy as little space as possible. After unfolding,
the mechanical components take the configuration shown in Fig. 26(a).
The SAR antenna consists in two identical subsystems, each with
three coupled four-bar links that unfold two panels on each side. The
central panel is attached to the main body of the satellite. Each unfold-
ing system has two degrees of freedom, driven individually by actuators
located in joints A and B. In the first phase of the unfolding process
the panel 3 is rolled out, around an axis normal to the main body, by
a rotational spring-damper-actuator in joint A, while the panel 2 is held
down by blocking the joints D and E. The second phase begins with
the joint A blocked, next the panels 2 and 3 are swung out to the final
position by a rotational spring-damped-actuator.
The model used for one half of the folding antenna, schematically
depicted Fig. 27, is composed of 12 bodies, 16 spherical joints and 3
revolute joints. The central panel is attached to the satellite, defined
as body 1, which has much higher mass and inertia. The data for this
antenna is reported in the work of Anantharamann [24].
Figure 27. The SAR antenna: a) half unfolded state b) folded antenna; c) multibody
model.
In the first phase of the unfolding antenna, the rotational spring-
damped-actuator is applied in the revolute joint R
3
. For the second
phase, the revolute joint R
3
is blocked and the system is moved to
the next equilibrium position by a spring-actuator-damped positioned
in joint R
1
. The unfolding processes for rigid and flexible models are
shown in Fig. 28, only for its first phase.
The different behavior between the rigid and the flexible models is
noticeable in Fig. 28. Though not shown here, the rotational actuator
moment responsible for the start of the unfolding is not correctly pre-
dicted by the rigid multibody model. Being a very light and flexible
structure, the discrepancies, if not identified during the design stage,
would lead to the failure of the unfolding process.
1.3 1.3
a) a)
Actuator ( Actuator (1) 1)
b) b)
Panel 3 (B3) Panel 3 (B3)
Panel 2 (B2 Panel 2 (B2))
Panel 1 (B1) Panel 1 (B1)
c) c) c) c)
84 ICTAM04
Figure 28. First phase of the unfolding of the SAR antenna (rigid and flexible
models).
Nonlinear Deformations in Multibody Systems
For flexible multibody systems experiencing nonlinear geometric and
material deformations, the equations of motion for a flexible body are
given by Eq. (29). However, due to the time variance of all its coefficients,
Eq. (29) is not efficient for computational implementation. Instead, by
considering a lumped mass formulation for the mass matrix and referring
the nodal accelerations to the inertial frame, the equations of motion for
a single flexible body take the form of [6]


mI +
¯
AM

¯
A
T

¯
AM

S 0

_
¯
AM

S
_
T
J

+S
T
M

S 0
0 0 M
ff


⎤⎡


¨r
˙ ω

¨ q

f



=


f
r
ff +
¯
AC

δ
n


˜
ω

J

ω

−S
T
C

δ

¯
I
T
C

θ
g

f
−f −(K
L
+K
NL
) u



(32)
where the absolute nodal displacements are written as
¨ q

kf

_
¨
d

¨ α

_
k
= ¨ u

k
+
_
A
T

_
˜ x
k
+
˙
δ
k
_

0 I
_
_
¨r
˙
ω

_
+
_
˜ ω

˜ ω

(x
k

k
)

+ 2˜ ω
˙
δ

k
˜ ω
˙
θ

k
_
(33)
with x
k
being the position of node k in the reference configuration. In
Eq. (32) M

is a diagonal mass matrix containing the mass of the n
boundary nodes,
¯
A
T
= [A. . . A]
T
, S =
_
_
˜ x

1
+
˜
δ

1
_
T
. . .
_
˜ x

n
+
˜
δ

n
_
T
_
T
Multibody Dynamics: Bridging for Multidisciplinary Applications 85
and
¯
I = [I . . . I]
T
where A is the transformation matrix from the body
fixed to global coordinate coordinates and x
k
denotes the position of
node k. Vectors C

δ
and C

θ
represent respectively the reaction force
and moment of the flexible part of the body over the rigid part, given
by
C

δ
= g

δ
−F
δ
−(K
L
+K
NL
)
δδ
δ

−(K
L
+K
NL
)
δθ
θ

,
C

δ
= g

θ
−F
θ
−(K
L
+K
NL
)
θδ
δ

−(K
L
+K
NL
)
θθ
θ

.
(34)
The coupling between the rigid body motion and the system deforma-
tions is fully preserved. For a more detailed description of the formula-
tion, and the notation, the interested reader is referred to reference [6].
As an application example of the nonlinear formulation for flexible
multibody systems, a sports vehicle with a front crash-box is analyzed
for various impact scenarios, represented in Fig. 29, where the angle of
Angle 10º
no friction
Angle 20º
frict ion = 0.5
Angle 10º
frict ion = 0.5
10 10 cm ramp
Angle 20º
no friction
(a) (e) (d)
Figure 29. Different impact scenarios for the sports vehicle.
Figure 30. Motion of the vehicle for a 20

oblique impact without contact friction
and for impact with an oblique surface for a vehicle traveling over a ramp.
86 ICTAM04
impact and the topology of the road are different. The simulations are
carried until the vehicle reaches a full stop.
The vehicle motion, for the oblique impact scenario presented in
Fig. 29, is characterized by a slight rotation of the vehicle during im-
pact. This rotation is more visible in the case of frictionless impact. At
the simulated impact speed the influence of the car suspension elements
on the deformation mechanism is minimal.
6. Conclusions
The multibody dynamics formalisms provide an extremely efficient
framework to incorporate different disciplines. The behavior of a good
number of phenomena in different problems can be represented by kine-
matic constraints (e.g., contact, muscle action, guidance) or by contact
forces (e.g, impact phenomena, control, general interactions). However,
different disciplines use different preferred numerical methods to solve
their equilibrium equations which lead to difficulties in the co-simulation
of different systems. The use of multibody formalisms in biomechanics
presents a strong increase due to the suitability to model contacts, mus-
cles, anatomical joints, data processing, etc. The treatment of structural
components with large rotations or of rotating bodies with structural de-
formations finds in the flexible multibody dynamics efficient methods to
deal with the problem. A continued effort to close the gap between the
flexible multibody dynamics and the nonlinear finite element method is
required. The need for more robust and efficient numerical methods to
handle the specific forms of the MBS equations and the discontinuities
associated to intermittent and ‘fast’ behaviors are still required.
Acknowledgements
The contents of this work result from a team effort and collaborations
with many co-workers among which the contribution by Miguel Silva,
Joao Gon¸ ˜ calves, Jo˜ ¸ ao Pombo, Manuel Seabra Pereira, Jo˜ ˜ ao Abrantes, ˜
Augusta Neto and Rog´erio Leal are gratefully acknowledged. ´
References
[1] P. Nikravesh, Computer-Aided Analysis of Mechanical Systems, Prentice-Hall,
Englewood Cliffs, New Jersey 1988.
[2] J. Garcia de Jalon, E. Bayo, Kinematic and Dynamic Simulation of Mechanical
Systems – The Real-Time Challenge, Springer-Verlag, Berlin 1994.
[3] P. Nikravesh and G. Gim, Systematic construction of the equations of motion
for multibody systems containing closed kinematic loops, Journal of Mechanical
Design, Vol. 115, No.1, pp.143–149, 1993.
Multibody Dynamics: Bridging for Multidisciplinary Applications 87
[4] C.W. Gear, Numerical solution of differential-algebraic equations, IEEE Tran-
sactions on Circuit Theory, Vol. CT-18, pp.89–95, 1981.
[5] J. Gon¸calves and J. Ambr´osio, Complex flexible multibody systems with applica- ´
tion to vehicle dynamics, Multibody System Dynamics, Vol. 6, No.2, pp.163–182,
2001.
[6] J. Ambr´osio and P. Nikravesh, Elastic-plastic deformation in multibody dyna- ´
mics, Nonlinear Dynamics, Vol. 3, pp.85–104, 1992.
[7] F. Pfeiffer and C. Glocker, Multibody Dynamics with Unilateral Contacts, John
Wiley and Sons, New York 1996.
[8] H. Lankarani and P. Nikravesh, Continuous contact force models for impact
analysis in multibody systems, Nonlinear Dynamics, Vol. 5, pp.193–207, 1994.
[9] M. Valasek, Z.
ˇ
Sika, Evaluation of dynamic capabilities of machines and robots,
Multibody System Dynamics, Vol. 5, pp.183–202, 2001.
[10] H. Møller and E. Lund, Shape Sensitivity Analysis of Strongly Coupled Fluid-
Structure Interaction Problems, [in:] Proc. 8th AIAA/USAF/NASA/ISSMO
Symposium on Multidisciplinary Analysis and Optimization, Long Beach, CA.
AIAA Paper No.2000–4823, 2000.
[11] H. Møller, E. Lund, J. Ambr´osio and J. Gon¸ ´ calves, Simulation of fluid loaded ¸
flexible multiple bodies, Multibody Systems Dynamics, Vol. 13, No.1, 2005.
[12] J. Pombo and J.Ambr´osio, General spatial curve joint for rail guided vehi- ´
cles: kinematics and dynamics, Multibody Systems Dynamics, Vol. 9, pp.237–264,
2003.
[13] J. Pombo and J. Ambr´osio, ´ A multibody methodology for railway dynamics ap-
plications, Technical Report IDMEC/CPM-04/002, IDMEC, Instituto Superior
T´ecnico, Lisboa, Portugal, 2004. ´
[14] M. Silva and J. Ambr´osio, Kinematic data consistency in the inverse dynamic ´
analysis of biomechanical systems, Multibody System Dynamics, Vol. 8, pp.219–
239, 2002.
[15] D. Tsirakos, V. Baltzopoulos and R. Bartlett, Inverse Optimization: Functional
and Physiological Considerations Related to the Force-Sharing Problem, Critical
Reviews in Biomedical Engineering, Vol. 25, Nos.4&5, pp.371–407, 1997.
[16] M. Silva and J. Ambr´osio, ´ Human Motion Analysis Using Multibody Dynamics
and Optimization Tools, Technical Report IDMEC/CPM-04/001, IDMEC, In-
stituto Superior T´ecnico, Lisboa, Portugal, 2004. ´
[17] G.T.Yamaguchi, Dynamic Modeling of Musculoskeletal Motion, Kluwer Acad-
emic Publishers, Boston 2001.
[18] H. Hertz Gesammelte Werke , Leipzig, Germany 1895.
[19] N.W. Murray, The static approach to plastic collapse and energy dissipation in
some thin-walled steel structures, [in:] Structural Crashworthiness, N. Jones and
T. Wierzbicki [eds.], pp.44–65, Butterworths, London 1983.
[20] C.M. Kindervater, Aircraft and helicopter crashworthiness: design and simula-
tion, [in:] Crashworthiness Of Transportation Systems: Structural Impact And
Occupant Protection, J.A.C. Ambrosio, M.S. Pereira and F.P Silva [eds.], NATO ´
ASI Series E. Vol. 332, pp.525–577, Kluwer Academic Publishers, Dordrecht
1997.
88 ICTAM04
[21] P.E. Nikravesh, I.S. Chung, and R.L. Benedict, Plastic hinge approach to vehicle
simulation using a plastic hinge technique, J. Comp. Struct. Vol. 16, pp.385–400,
1983.
[22] 30 mph Rollover Test of an AM General Model M151-A2 1/4 Ton Jeep, The
Transportation Research Center of Ohio, Test Report, 1985.
[23] K.-J. Bathe and S. Bolourchi, Large displacement analysis of three-dimensional
beam structures, Int. J. Num. Methods in Engng., Vol. 14, pp.961–986, 1979.
[24] M. Anantharaman and M. Hiller, Numerical simulation of mechanical systems
using methods for differential-algebraic equations, Int. J. Num. Meth. Eng.,
Vol. 32, pp.1531–1542, 1991.

62

ICTAM04

Figure 1. Natural biological and artificial engineering systems for which multibody dynamics provides irreplaceable modeling methodologies.

deal with large overall motion. Therefore, it is no surprise that many of the most recent formulations on flexible multibody dynamics and on finite element methods with large rotations share some common features. In multibody dynamics methods, the body-fixed coordinate frames are generally adopted to position each one of the system components and to allow for the specification of the kinematic constraints that represent the restrictions on the relative motion between the bodies. Several formalisms are published suggesting the use of different sets of coordinates, such as Cartesian [1], natural [2] and relative coordinates [3]. Depending on the type of applications, each of these types of coordinates has advantages and disadvantages. Due to the ease of the computational implementation, their physical meaning and the widespread knowledge of their features, all the formalisms presented in this work are based on the use of Cartesian coordinates. The methodological structure of the equations of motion of the multibody system obtained allows the incorporation of the equilibrium equations of a large number of disciplines and their solution in a combined form. The description of the structural deformations exhibited by the system components by using linear [5] or non-linear finite elements [6] in the framework of multibody dynamics is an example of the integration of the equations of equilibrium of different specialties. Of particular importance for the applications pursued with the methodologies proposed is the treatment of contact and impact, which is introduced in the multibody systems equations by using either unilateral constraints

Multibody Dynamics: Bridging for Multidisciplinary Applications 63 [7] or a continuous contact force model [8]. i (1) According to this definition. are of importance [10. qT ]T . (3) with respect to time yields: ¨ Φ (q. q. . are defined through the introduction of kinematic relationships written as [1]: Φ (q. The position and orientation of each body i in the space is described by a position vector ri and a set of rotational coordinates pi . 2. 1 2 nb (2) The dependencies among system coordinates. 11]. which result from the existence of mechanical joints interconnecting several bodies. automotive and railway dynamics are used to demonstrate the developments reviewed here. (3) where t is the time variable. qT . The system kinematic constraints are added to the equations of motion using the Lagrange multipliers technique [1]. Application cases involving the modeling of realistic mechanisms. a multibody system with nb bodies is described by a set of coordinates in the form: q = [qT . The second time-derivative of Eq. . The coupling between the fluid and structural dynamics equations allows for the development of applications. q. biomechanics or robotics and their integration with the multibody equations [9]. t) = 0. t) = 0 ≡ D¨ = γ. ˙ ¨ q (4) where D is the Jacobian matrix of the constraints. q is the acceleration ¨ vector and γ is the vector that depends on the velocities and time. especially for cases with large absolute or relative rotations in the system components. passive safety of road and rail vehicles. which are organized in a vector as [1]: qi = [rT . which is used only for the driving constraints. where the fluid-structure interaction is analyzed. pT ]T . The research carried at IDMEC provides the examples offered in this work. . impact and human locomotion biomechanics. The availability of the state variables in the multibody formulation allows for the use of different control paradigms in the framework of vehicle dynamics. . Rigid Multibody Dynamics A multibody system is defined as a collection of rigid and/or flexible bodies constrained by kinematic joints and eventually acted upon by a set of internal and/or external forces. Denoting by λ the .

is outlined in Fig. Solution of the forward dynamic analysis of a multibody system. . The solution procedure starts by the determination of the initial positions and velocities of the system components. represented by Eq. Forward Dynamics The computational strategy used to solve the forward dynamics of the system. are physically related with the reaction forces generated between the bodies interconnected by kinematic joints. are calculated and assembled in the equations of motion. Equation (5) is then solved to find the system accelerations. the equations of motion for a mechanical system are written as M DT D 0 q ¨ λ = f γ (5) where M is the global mass matrix. q (6) The usual procedures to handle the integration of sets of differentialalgebraic equations must still be applied in this case in order to eliminate the constraint drift or to maintain it under control [1. and f is the force vector. at time t. the Jacobian matrices.64 ICTAM04 vector of the unknown Lagrange multipliers. associated with the kinematic constraints. the new positions and velocities for time t + ∆t are calculated by using a variable Figure 2. containing the mass and moments of inertia of all bodies. given by [1] f (c) = −ΦT λ. By integrating the current velocities and the system accelerations. 2]. 2. The Lagrange multipliers. the forces and the right-hand-side of the kinematic acceleration constraint equations vectors. (5). and in the process the Lagrange multipliers. Next. the system inertia. containing all forces and moments applied to the system bodies plus the gyroscopic forces.

The path is defined by a parametric curve g(L). Throughout this work it is demonstrated that all engineering applications foreseen here are implemented. not only its path has to be followed. (5). The kinematic constraint is Φ(pmc. according to spatial characteristics of the curve. z). ri is the vector that defines the location of the body-fixed coordinate system (ξ. using the moving Frenet frame associated with the track centerline based on the work by Pombo and Ambrosio [12]. variable time-step integration procedure [1. The forward dynamics simulation proceeds until the previously set final time is reached. located on a rigid body i. that has to follow the specified path depicted in Fig. ζ)i .Multibody Dynamics: Bridging for Multidisciplinary Applications 65 order. η. 4]. 2 is used in general purpose multibody dynamics codes. Local frame alignment constraint. Ai is the transformation matrix from the body i fixed coordinates to the global reference frame. such as DAP-3D [1]. When a body travels along a guide. The procedure outlined in Fig. 3. . which is controlled by a global parameter L that represents the length travelled along the curve until the current location of point R. is outlined ´ next. Prescribed Motion Constraint. The objective here is to define the constraint equations that enforce that a point of a rigid body follows the reference path [12]. y. i (7) where rR = ri +Ai s R represents the coordinates of point R with respect i i to the global coordinate system (x. and s R represents the coordinates of point R with rei x Figure 3. Consider a point R. Application Example of a Roller Coaster.3) = 0 ≡ rR − g(L) = 0. but also its spatial orientation has to be prescribed. either by developing specific kinematic constraints or by implementing force models in Eq. The formulation adopted to implement these kinematic constraints. 2.

.. (8) ⎭ ⎩ ⎭ ⎩ T c n · uζ This kinematic constraint ensures that the alignment remains constant throughout the analysis. The transformation matrix from the body i fixed coordinates to the global coordinate system is: Ai = [uξ uη uζ ]i (9) defining the following unit vectors as: u1 = {1 0 0}T . 4.o. represented in Fig.66 ICTAM04 spect to the body-fixed reference frame. corresponding to 3 car bodies. uη . uη .3) = 0 ≡ bT · uξ − b = 0. b)L . uζ )i represent the unit vectors associated with the axis of the body-fixed coordinate system (ξ. ⎭ ⎩ ⎭ ⎩ T c n Ai u3 which constitutes the second part of the path following constraint. (10) Equation (8) is now rewritten in a more usable form as: ⎫ ⎧ ⎫ ⎧ T ⎨n Ai u1 ⎬ ⎨a⎬ Φ(lf ac. η. A view of the motion of the roller coaster is displayed in Fig. For notational purposes (·) means that (·) is expressed in body-fixed coordinates. n. 6 wheelsets and 2 connection bars. The roller coaster vehicle consists of a train with three cars that are interconnected by linking bars. The complete vehicle model only has 1 d. ζ)i . uζ )i and the local frame (t. (11) Roller-Coaster Dynamics. The multibody model of the vehicle is assembled using eleven rigid bodies. Let the Frenet frame of the general parametric curve g(L) be defined by the principal unit vectors (t. n. (5). Consider that (uξ . 6 and the details of the analysis are found in reference [13]. which is the longitudinal motion of the cars. The motion of the vehicle is guided by the dynamics described by Eq. 5. The second part of the constraint ensures that the spatial orientation of body i remains unchanged with respect to the moving frame of the reference path represented in Fig. At the initial time of analysis. 3.3) = 0 ≡ bT Ai u1 − b = 0. The path-following constraint is used to enforce the vehicles to follow the rail for the prescribed geometry. u2 = {0 1 0}T . Let the roaller-coaster rail be defined with the spatial geometry described in Fig. the relative orientation between the body vectors (uξ .f. b)L leads to ⎫ ⎧ ⎫ ⎧ T ⎨n · uξ ⎬ ⎨a⎬ Φ(lf ac. u3 = {0 0 1}T .

the only unknowns are the internal forces. (5) be re-written as M¨ + DT λ = g q (12) . Note that the study of these vehicles only requires the use of the pathfollowing constraint. Therefore. Let the first row of Eq. Snap shots of the roller coaster motion as observed from the second car. The contact forces are not explicitly used but they can be calculated from the Lagrange multipliers associated to the path constraint. Multibody model of the roller coaster vehicle. Figure 6. 3 6 10 7 11 x 1 2 4 5 8 9 Figure 5. z View of the roller coaster as used in the simulation. Inverse Dynamics In many applications all external forces are known and the motion of the system is also known.Multibody Dynamics: Bridging for Multidisciplinary Applications 67 Figure 4.

. The reaction forces at the joints are given by: g(c) = −DT λ. 8 for the knee joint. that develop for known motions. muscle and anatomical joint reaction forces.e. Application to Biomechanics: Gait Analysis For biomechanical applications in gait a three-dimensional model. 7. joint actuators such as the one represented in Fig. To drive the biomechanical model in the inverse dynamic analysis. The actuators are the kinematic constraints in which the angle between two adjacent segments is a known function of time. The biomechanical model. presented in Fig. i. is used [14]. interconnected by revolute and universal joints. (14) The solution of the equations of motion in inverse dynamics can be used to solve for the internal forces of the human body. It has a kinematic structure made of thirtythree rigid bodies.68 which is q DT λ = g − M¨. These additional equations are added to the system kinematic equation so that the number of . in such a way that sixteen anatomical segments are represented. its kinematic structure and a detail of the ankle joint. are specified. Joint Moments-of-Force: A Determinate Problem. 16 11 1 3 v20 v20 5 6 31 19 33 12 3 4 4 2 2 32 25 24 v16 v15 23 v16 26 27 6 13 v14 20 v15 v13 22 v17 28 v17 29 v19 9 30 1 7 1 9 14 v12 17 18 v12 v13 21 v18 v21 1 v22 5 7 12 2 10 15 v11 16 v11 v4 5 6 7 8 8 8 10 13 v9 v4 11 14 v10 v5 v5 9 10 4 v3 3 v3 v6 11 v6 12 v8 v7 13 9 14 15 16 v1 1 2 v2 15 Figure 7. ICTAM04 (13) Equation (13) emphasizes that the only unknowns of the system are the Lagrange multipliers.

i = 1. . is totally determined. The optimization problem is stated as: minimize F0 (ui ) ⎧ ⎪ fj (ui ) = 0. j = (nec + 1) . By using optimization techniques to find the muscle forces that minimize a prescribed objective function. Joint actuator associated with the knee joint and muscle actuator. ⎨ subject to f (u ) 0. . . . Several cost functions have been proposed for the study of the redundant problem in biomechanics [15]. nsv (15) where ui are the state variables bounded respectively by ulower and uupper . nec . The minimization of the sum of the cube of the muscle stresses [16] is often used in applications involving human . . The inverse dynamics problem.. The solution of the inverse dynamics problem with muscle actuators introduces indeterminacy in the biomechanical system. The minimization of the cost functions simulate the physiological criteria adopted by the central nervous system when deciding which muscles to recruit and what level of activation to obtain the adequate motion. . since it involves more unknowns than equations of motion. . F0 (ui ) is the objective or cost function to minimize and fi (ui ) are constraint equations that restrain the state variables..Multibody Dynamics: Bridging for Multidisciplinary Applications 69 O m1 m1 Im1 = O m2 m2 Im2 = 2 I m3 O m3 m3 Figure 8. a solution for the problem is obtained. representing the net moments-of-force of the muscles that cross those joints. . ntc . as stated here. Equation (13) is solved to obtain the Lagrange multipliers associated with the joint actuators. ⎪ j i ⎩ lower ui ui uupper i j = 1.. Muscle Forces: A Redundant Problem. non-redundant constraint equations becomes equal to the number of coordinates. .

Muscle forces for the hamstrings and triceps surae.70 ICTAM04 Figure 9. Figure 10. Lower extremity muscle apparatus. .

(19) k n rik = ( ik )n n rik i rli l k* l t rik = rik − ( ik i )n r jl j j rij Figure 11. 9. represented in Fig. be defined by points i. with which a thickness e is associated. 3. The position of the point k with respect to point i of the surface is rik = rk − ri rt = rik − rT n n. 11 will impact. is modeled having the muscles with the physiological data described in Yamaguchi [17]. where point k of the body shown in Fig. Contact detection between a finite element node and a surface. 10. . but not through its ‘back’ surface.39 N/cm locomotion apparatus. To illustrate the type of results obtained for the muscle forces in a case of normal cadence gait of a 50%ile male. the muscle forces for the hamstrings and triceps surae are presented in Fig. given by rn = rT n n. ik ik (17) which is decomposed in a tangential and a normal component. The human muscle strength with a constant value of 31. ik ik (18) The necessary conditions for contact are that node k penetrates the ‘front’ surface of the patch. The state variables associated with muscle actuators represent muscle activations that can only assume values between 0 and 1.Multibody Dynamics: Bridging for Multidisciplinary Applications 71 locomotion F0 = nma σ ¯ m=1 Flm Fl˙m m2 F0 3 a m (16) ¯ where nma are the number of muscle actuators and σ is the specific 2 [17]. Contact and Impact Let a triangular patch. These conditions are written as 0 rT n ik e. The normal to the outside surface of the contact patch is defined as n = rij × rjl . j and l.

72 ICTAM04 The remaining necessary conditions for contact result from the need for the node to be inside of the triangular patch. the normal components of the node k velocity and acceleration. Continuous Contact Force Model An alternative description of contact considers this to be a continuous event where the contact force is a function of the penetration between the surfaces. For flexible bodies let us assume a fully plastic nodal contact. This leads to the continuous force contact model. These three extra conditions are ˜t rij rik T n 0. proposed by Lankarani and Nikravesh [8]. However. the change of sign of the force is in fact the change of sign of the multiplier. depending on the contact force model actually used. a kinematic constraint is imposed.i = Kδ n + Dδ u (23) .. The kinematic constraint implied by Eqs. qk = q k ¨ ¨ (−) − qk ¨ (−)T n n (21) where qk and qk represent the nodal velocity and acceleration imme˙ ¨ diately before impact respectively . with respect to the surface.. Let the contact force between two bodies be written as ˙ fs.e. This contact model is not suitable to be used directly with rigid body contact. Other forms of this contact model can be found in the work by Pfeiffer and Glocker [7].e. (21) is removed when the normal reaction force between the node and the surface becomes opposite to the surface normal. i. ˜t rjl rik T n 0 and ˜t rki rik T n 0. n fT fk = −fk n > 0. (20) Equations (19) and (20) are necessary conditions for contact. (−) (22) It should be noted that the contact force is related to the Lagrange multiplier associated by the kinematic constraint defined by Eqs. Unilateral Constraints If contact between a node and a surface is detected. i. The sudden change of the rigid body velocity and acceleration would imply that the velocity and acceleration equations resulting from the kinematic constraints would not be fulfilled. are null during contact ˙ qk = q k ˙ (−) (−) − qk ˙ (−)T n n. they may not be sufficient to ensure effective contact. (22). Therefore. and briefly described here.

In a general case of a railway vehicle one or two points of each wheel are in contact with the rail. Let the generalized geometry of the rail and wheel be described by generalized surfaces resulting from sweeping the rail profile along the rail Fla (Le ct ct) Tre Tre con Figure 12. The possibility of detecting contact in different diametric sections allows predicting derailment and it is. D is a damping coefficient and u is a unit vector normal to the impacting surfaces. The diametric section that contains the wheel flange contact point makes an angle sf Rw with the diametric section that contains the wheel tread contact point. represented in Fig. methodologies that provide accurate models to represent the phenomena are of particular importance.i = Kδ n 1 + ˙ 3 1 − e2 δ u ˙ 4 δ (−) (24) ˙ where δ (−) is the initial contact velocity and e is the restitution coefficient. 12. The stability of the running vehicle depends ultimately on the rail-wheel contact and on the vehicle primary suspension. δ is the pseudo-velocity of penetration. therefore. Using the hysteresis dissipation model and the equivalent stiffness. Therefore.Multibody Dynamics: Bridging for Multidisciplinary Applications 73 ˙ where δ is the pseudo-penetration. Note that K is a function of the geometry and material properties of the impacting surfaces. Two points of contact in the rail and wheel surfaces: lead contact. . calculated for instance for Hertzian elastic contact [18]. the nonlinear contact force is fs. as shown in Fig. of utmost importance. K is the equivalent stiffness. Application to Railway Dynamics – The Wheel-Rail Contact Problem One of the interesting applications of multibody dynamics with contact mechanics is the description of the wheel-rail contact in railway dynamics. 12.

This condition means that nj has null projections over the tangent vectors tu and tw : i i ⎧ ⎨nT tu = 0. j i (25) nj × ni = 0 ⇔ ⎩nT tw = 0. This system of equations provides solutions for the location of the candidates to contact points that have to be sorted out. x Figure 13. 13. The geometric conditions for contact between the convex surfaces are defined by vector products defined between the surfaces. In order to ensure that the search for the contact points is between convex surfaces. Candidates to contact points between two parametric surfaces. This condition is mathematically written as: ⎧ ⎨dT tu = 0. i (26) d × ni = 0 ⇔ ⎩dT tw = 0. the wheel profile is divided in treat and flange profiles. The first condition is that the surfaces normals ni and nj at the candidates to contact points have to be parallel. The contact between the rail and one of the wheel surfaces is described generically in Fig. . s and t that define the two surfaces. which represents the distance between the candidates to contact points. has to be parallel to the normal vector ni . i The geometric conditions (25) and (26) provide four nonlinear equations with four unknowns. the four parameters u.74 ICTAM04 centerlines and the wheel profiles around the base circle of the wheel. w. j i The second condition is that the vector d. where the mating surfaces are represented as free surfaces.

14. and it is used hereafter [13]. developed in the work by Pombo [13]. The multibody model of the trailer vehicle of the train. Figure 15. dampers and rigid connecting ele- Figure 14. shown in Fig. 15. This assembly of connective elements constitutes the secondary suspension. also achieved by another set of springs. the Polach formulation or the Heuristic nonlinear creep model. sketched in Fig. which is the main one responsible for the passenger’s comfort. the normal force is calculated using Eq. is composed of the car shell suspended by a set of springs. Schematic representation of the ML95 trainset.Multibody Dynamics: Bridging for Multidisciplinary Applications 75 The coordinates of the candidates to contact points are determined by solving an optimal problem and the distance between such points is calculated in the process. which is used by the Lisbon subway company (ML) for passengers’ traffic. (24) and the tangential forces are evaluated using the Kalker theory. The connections between the bogies chassis and the wheelsets. . The wheel-rail contact model outlined here is used to model the ML95 trainset. It has been found that the Polach formulation provides the best approach for the tangent forces. (27) When contact is detected. The points are in contact if dT nj 0. Secondary suspension model of the ML95 trailer vehicle. dampers and other rigid connecting elements on the bogies.

running in a circular track with a radius of 200 m with velocities of 10 and 20 m/s. derailment does not occur and the analysis proceeds up to end. Lift of the right wheel of the leading wheelset for vehicle forward velocities of 10 and 20 m/s. The primary suspension is the main suspension responsible for the vehicle running stability. 16. ments. flange contact occurs. 18 for a vehicle forward velocity of 10 m/s and .76 Bogie frame Axlebox ICTAM04 Three-dimensional spring-damper elements a) Wheelset b) Figure 16. which makes the Kalker linear theory inappropriate to compute the creep forces. Nevertheless. Therefore only the Heuristic and the Polach creep force models must be considered. Another aspect to note is that flange contact is detected with all creep force models. constitute the primary suspension represented in Fig. 17 shows that contact forces obtained with the Kalker linear theory originate the lift of the outer wheel of the front wheelset at the entrance of the curve. b) Suspension model with springs and dampers. Despite this wheel lift. The simulation results of the vehicle. Primary suspension model of the ML95 trailer bogie: a) Threedimensional spring-damper elements. Figure 17. such results are not realistic since the existence of flange contact involves high creepages. where the centrifugal forces effect is balanced by the track cant. show that the prediction of flange contact is of fundamental importance. Even when running at the speed of 10 m/s. Lateral flange forces develop on the wheels of both wheelsets of the front bogie as presented in Fig. Fig. using the Kalker linear theory.

During curve negotiation.Multibody Dynamics: Bridging for Multidisciplinary Applications 77 using the Polach creep force model. using the Polach creep force model. when running at 20 m/s. Flange contact Front wheelset (Ws 4) Flange contact Rear wheelset (Ws 3) Figure 19. This is explained by the fact that. For the velocity of 20 m/s. for the velocity of 10 m/s. For structural crashworthiness it is 25 000 Le ft W s 3 ( P ol ach) 20 000 R ig ht W s 3 ( P olach ) Le ft W s 4 ( P ol ach) La te ra l Flang e F orc e [N ] 15 000 R ig ht W s 4 ( P olach ) 10 000 5 000 0 -5 0 0 0 0 3 6 9 12 15 18 21 24 27 30 T im e [s ] Figure 18. Lateral flange forces on the wheels of both wheelsets of the front bogie for a vehicle forward velocity of 10 m/s. Contact configuration during curve negotiation. the vehicle negotiates the curve with a velocity higher than the balanced speed. Flexible Multibody Dynamics with Plastic Hinges Many applications of multibody dynamics require the description of the flexibility of its components. the flange contact occurs on the outer and in the inner wheels of the vehicle. the outer wheel of the leading wheelset and the inner wheel of the rear wheelset have permanent flange contact. 19. 4. only the outer wheels have flange contact. . Referring to Fig.

which makes the procedure adequate for the early phases of vehicle design.07V00. These deformations. Multibody models obtained with this method are relatively simple. load application points. The plastic hinge constitutive equation can be modified to account for the strain rate sensitivity of some materials. called plastic hinges. The methodology described herein is known in industry as conceptual modeling [20]. The use of multibody dynamics with plastic hinges is an alternative formulation that allows building insightful models for crashworthiness. the individual structural members are overloaded giving rise to plastic deformations in highly localized regions. respectively. Localized deformations on a beam and a plastic hinge. A dynamic correction factor is used to account for the strain rate sensitivity given by [21]. P V Pd /Ps = 1 + 0. presented in Fig.82 . The characteristics of the spring-damper that describes the properties of the plastic hinge are obtained by experimental component testing.78 ICTAM04 often unfeasible to use large nonlinear finite element models. (28) where Pd and Ps are the dynamic and static forces. as illustrated in Fig. joints or locally weak areas [19]. finite element nonlinear analysis or simplified analytical methods. Figure 20. The plastic hinge concept has been developed by using generalized spring elements to represent constitutive characteristics of localized plastic deformation of beams and kinematic joints to control the deformation kinematics [21]. 21. develop at points where maximum bending moments occur. The force or moment . Formulation of Plastic Hinges In many impact situations. 20 for structural bending. and V0 is the relative velocity between the adjacent bodies.

with a 50%tile.05 . The three occupants. to apply due to the plastic hinge is multiplied by the ratio calculated in Eq. integrated in the vehicle are seated.15 . The biomechanical models for the occupants are similar to those described in Fig. is used to demonstrate the plastic hinge approach to complex crash events. and a tire model [16].Multibody Dynamics: Bridging for Multidisciplinary Applications 60 79 0 Moment ( kNm) 15 30 45 Analytical Test 0 . .10 . Two occupants in the front of the vehicle have shoulder and lap Figure 22. 7. (28) before it is used in the force vector of the multibody equations of motion. The model includes all moving components of the vehicle. suspension systems and wheels. shown in Fig. Initial position of the vehicle and occupants for the rollover. Application of the Plastic Hinge Approach to Crashworthiness of Surface Vehicles The multibody of an off-road vehicle with three occupants.20 . 22.25 Rotation ( Rad ) Figure 21. Plastic hinge bending moment and its constitutive relationship.

The outcome of the experimental test. The results of the simulation are pictured in Fig. View of the experimental test for the truck rollover. occupants.41 m/s until the impact with a waterfilled decelerator system occurs. The occupants in the front of the vehicle are held in place by the seat belts. At this point the rear occupant is ejected. 24. Views of the outcome of the rollover simulation of a vehicle with three Figure 24.80 ICTAM04 seat belts while the occupant seated in the back of the vehicle has no seatbelt. The rollover situation for the simulation is such that the initial conditions correspond to experimental conditions where the vehicle moves on a cart with a lateral velocity of 13. Bouncing from the inverted position. is further used to validate the vehicle model [21]. mainly due to the ground – tire contact friction forces. the vehicle impacts the ground with its rollbar cage. The vehicle is then ejected with a roll angle of 23 degrees. Upon continuing its roll motion. the vehicle completes another half turn and impacts the ground with the tires again. Figure 23. An experimental test of the vehicle was carried out at the Transportation Research Center of Ohio [22]. being an overview of the footage obtained shown in Fig. The vehicle first impacts the ground with its left tires. The HICs for all occupants largely exceed 1000. 23 by several frames of the animation. The rollover motion of the vehicle proceeds with an increasing angular velocity. while the ejection of the rear occupant is complete. which is rather similar to the outcome of the simulation. . which indicates a very high probability of fatal injuries for the occupants under the conditions simulated.

General motion of a flexible body. Let the finite element method be used to represent the equations of motion of the flexible body. t ζ t η t b ∆t ζ t ∆ ted updated ation 0 0 ζ t ′ η b t t h ∆ b Figure 25. The local coordinate frame ξηζ attached to the flexible body. 25.Multibody Dynamics: Bridging for Multidisciplinary Applications 81 5. . the assembly of all finite elements used in the discretization of a single flexible body results in its equations of motion written as [6] ⎤⎡ ⎤ ⎡ ⎤ ⎡ ⎤ ⎡ ⎤ gr ¨ r sr 0 Mrr Mrf Mrf ⎣Mφr Mφφ Mφf ⎦ ⎣ω ⎦ = ⎣g φ ⎦ − ⎣s φ ⎦ − ⎣0⎦ ˙ Mf r Mf φ Mf f u ¨ gf sf f ⎤⎡ ⎤ ⎡ 0 0 0 0 ⎦ ⎣ 0 ⎦ (29) 0 − ⎣0 0 u 0 0 KL + KN L ⎡ where ¨ and ω are respectively the translational and angular accelerar ˙ tions of the body-fixed reference frame and u denotes the nodal accele¨ rations measured in body fixed coordinates. is used to represent the gross motion of the body and its deformation. Referring to Fig. Flexible Multibody Dynamics with Finite Elements General Equations of Motion of a Single Body Let the principle of the virtual works be used to express the equilibrium of the flexible body in the current configuration t+∆t and let an updated Lagrangean formulation be used to obtain the equations of motion of the flexible body [23].

82 ICTAM04 Linear Deformations of Flexible Bodies In many situations it is enough to consider that the components of the multibody system experience only linear elastic deformations. The methodology is demonstrated through the application to the simulation of the unfolding of a satellite antenna. Furthermore. The reduced equations of motion for a linear flexible body are [5] Mr XT Mf r Mrf X I qr ¨ w ¨ = gr X T gf − sr XT sf − 0 Λw (31) where Λ is a diagonal matrix with the squares of the natural frequencies associated with the modes of vibration selected. the flexible part of the body is described by a sum of selected modes of vibration as (30) u = Xw where the vector w represents the contributions of the vibration modes towards the nodal displacements and X is the modal matrix. Due to the assumption of linear elastic deformations the modal matrix is invariant. Due to the reference conditions. the modes of vibration used here are constrained modes. For a more detailed discussion on the selection of the modes used the interested reader is referred to [5]. assume that the mode superposition technique can be used. the Synthetic Aperture Radar (SAR) antenna that is a part of the European research satellite Figure 26. Then. . The European satellite with the folded and unfolded configurations of the antenna.

The unfolding processes for rigid and flexible models are shown in Fig. The SAR antenna: a) half unfolded state b) folded antenna. 26. the revolute joint R3 is blocked and the system is moved to the next equilibrium position by a spring-actuator-damped positioned in joint R1 . 27. 28. The central panel is attached to the satellite. the mechanical components take the configuration shown in Fig. The second phase begins with the joint A blocked. next the panels 2 and 3 are swung out to the final position by a rotational spring-damped-actuator. while the panel 2 is held down by blocking the joints D and E. would lead to the failure of the unfolding process. defined as body 1. represented in Fig. For the second phase. 26(a). Being a very light and flexible structure.3 Actuator (1) (1) Panel 1 (B1) a) b) c) Figure 27. schematically depicted Fig. model. . c) multibody In the first phase of the unfolding antenna. the discrepancies. driven individually by actuators located in joints A and B. Panel 3 (B3) Panel 2 (B2) (B2) 1. After unfolding. The data for this antenna is reported in the work of Anantharamann [24]. 28. In the first phase of the unfolding process the panel 3 is rolled out. the rotational springdamped-actuator is applied in the revolute joint R3 . by a rotational spring-damper-actuator in joint A. The model used for one half of the folding antenna. only for its first phase. 16 spherical joints and 3 revolute joints. During the transportation the antenna is folded. if not identified during the design stage. is composed of 12 bodies. The SAR antenna consists in two identical subsystems. around an axis normal to the main body. in order to occupy as little space as possible. which has much higher mass and inertia.Multibody Dynamics: Bridging for Multidisciplinary Applications 83 ERS-1. Though not shown here. each with three coupled four-bar links that unfold two panels on each side. the rotational actuator moment responsible for the start of the unfolding is not correctly predicted by the rigid multibody model. The different behavior between the rigid and the flexible models is noticeable in Fig. Each unfolding system has two degrees of freedom. The central panel is attached to the main body of the satellite.

¯ AT = [A . S= ˜ x1 + δ 1 ˜ T ˜ . . the equations of motion for a flexible body are given by Eq. . . (32) M∗ is a diagonal mass matrix containing the mass of the n boundary nodes. In Eq. First phase of the unfolding of the SAR antenna (rigid and flexible Nonlinear Deformations in Multibody Systems For flexible multibody systems experiencing nonlinear geometric and material deformations. . by considering a lumped mass formulation for the mass matrix and referring the nodal accelerations to the inertial frame. Instead. A]T .84 ICTAM04 Figure 28. the equations of motion for a single flexible body take the form of [6] ⎡ ¯ ¯ mI + AM∗ AT ⎣ − AM∗ S T ¯ 0 ⎤⎡ ⎤ ¯ −AM∗ S 0 ¨ r T M∗ S ⎦ ⎣ω ⎦ ˙ J +S 0 qf ¨ 0 Mff ⎤ ⎡ ¯ fr + AC δ ˜ = ⎣n − ω J ω − ST C δ − ¯T C θ ⎦ (32) I g f − f − (KL + KN L ) u where the absolute nodal displacements are written as qkf ≡ ¨ ¨ d α ¨ = uk + ¨ k AT 0 ˙ − xk + δ k ˜ I + ¨ r ˙ ω (33) ω ˙ ω ω (xk + δ k ) + 2˜ δ k ˜ ˜ ωθk ˜ ˙ with xk being the position of node k in the reference configuration. (29) is not efficient for computational implementation. xn + δ n ˜ T T . (29). models). However. Eq. due to the time variance of all its coefficients.

. Motion of the vehicle for a 20◦ oblique impact without contact friction and for impact with an oblique surface for a vehicle traveling over a ramp. . Vectors C δ and C θ represent respectively the reaction force and moment of the flexible part of the body over the rigid part. Figure 30.5 Angle 10º friction = 0. represented in Fig. I]T where A is the transformation matrix from the body I fixed to global coordinate coordinates and xk denotes the position of node k. and the notation. Different impact scenarios for the sports vehicle. a sports vehicle with a front crash-box is analyzed for various impact scenarios.5 10 cm ramp (a) (d) (e) Figure 29. where the angle of Angle 10º no friction Angle 20º no friction Angle 20º friction = 0. 29. given by C δ = g δ − Fδ − (KL + KN L )δδ δ − (KL + KN L )δθ θ . . For a more detailed description of the formulation. C δ = g θ − Fθ − (KL + KN L )θδ δ − (KL + KN L )θθ θ . As an application example of the nonlinear formulation for flexible multibody systems.Multibody Dynamics: Bridging for Multidisciplinary Applications 85 and ¯ = [I . (34) The coupling between the rigid body motion and the system deformations is fully preserved. the interested reader is referred to reference [6].

143–149. Acknowledgements The contents of this work result from a team effort and collaborations with many co-workers among which the contribution by Miguel Silva. anatomical joints. E. Berlin 1994. Journal of Mechanical Design. etc. ´ e References [1] P.1. Prentice-Hall. muscle action. control. The simulations are carried until the vehicle reaches a full stop. muscles. Springer-Verlag.86 ICTAM04 impact and the topology of the road are different. . contact. At the simulated impact speed the influence of the car suspension elements on the deformation mechanism is minimal. Nikravesh and G. Vol. Kinematic and Dynamic Simulation of Mechanical Systems – The Real-Time Challenge. Jo˜ Pombo. Joao Gon¸ ˜ calves. 115. Bayo. [3] P. 1993. data processing. is characterized by a slight rotation of the vehicle during impact. pp. The vehicle motion. The use of multibody formalisms in biomechanics presents a strong increase due to the suitability to model contacts. Garcia de Jalon. The treatment of structural components with large rotations or of rotating bodies with structural deformations finds in the flexible multibody dynamics efficient methods to deal with the problem. No. However. The need for more robust and efficient numerical methods to handle the specific forms of the MBS equations and the discontinuities associated to intermittent and ‘fast’ behaviors are still required. New Jersey 1988. [2] J. Jo˜ Abrantes. for the oblique impact scenario presented in Fig.g. Englewood Cliffs. This rotation is more visible in the case of frictionless impact. The behavior of a good number of phenomena in different problems can be represented by kinematic constraints (e. A continued effort to close the gap between the flexible multibody dynamics and the nonlinear finite element method is required.. Systematic construction of the equations of motion for multibody systems containing closed kinematic loops.g. 6. different disciplines use different preferred numerical methods to solve their equilibrium equations which lead to difficulties in the co-simulation of different systems. impact phenomena. ¸ ˜ ao ao ˜ Augusta Neto and Rog´rio Leal are gratefully acknowledged. guidance) or by contact forces (e. Gim. 29. Manuel Seabra Pereira. Computer-Aided Analysis of Mechanical Systems. Nikravesh. Conclusions The multibody dynamics formalisms provide an extremely efficient framework to incorporate different disciplines. general interactions).

2001.237–264. Silva and J. Inverse Optimization: Functional and Physiological Considerations Related to the Force-Sharing Problem. 2004. Dordrecht 1997. E. pp.219– 239. Ambr´sio and J. Møller and E.P Silva [eds.C. pp. Multibody Dynamics with Unilateral Contacts. pp. 25.85–104. Boston 2001. Portugal. Pereira and F.183–202. Multibody System Dynamics. 2000. New York 1996. A multibody methodology for railway dynamics ap´ o plications. Bartlett. Ambr´sio. Lisboa. Leipzig. IDMEC. 2003.163–182.W. Vol.Yamaguchi. NATO ´ ASI Series E. ˇ [9] M.1. Gear. Z. CT-18. 9. Ambr´sio. Nonlinear Dynamics. CA. pp. Aircraft and helicopter crashworthiness: design and simulation.Multibody Dynamics: Bridging for Multidisciplinary Applications 87 [4] C. Instituto Superior T´cnico. . Evaluation of dynamic capabilities of machines and robots. [in:] Proc. 2001. John Wiley and Sons. [5] J. Critical Reviews in Biomedical Engineering. 8th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization. Valasek. IEEE Transactions on Circuit Theory. Gon¸ ´ o ¸ calves. Vol. [8] H. General spatial curve joint for rail guided vehi´ o cles: kinematics and dynamics. London 1983. Multibody Systems Dynamics. Vol. pp. No. Lankarani and P. pp. N. Pfeiffer and C. 2004. pp. Complex flexible multibody systems with applicac ´ o tion to vehicle dynamics. Jones and T. [10] H. J. Nikravesh. Murray. Dynamic Modeling of Musculoskeletal Motion. ´ e [14] M.2. Kluwer Academic Publishers. pp. Baltzopoulos and R. Butterworths.89–95. [16] M. Kluwer Academic Publishers. Multibody System Dynamics. Gon¸alves and J. [11] H. Shape Sensitivity Analysis of Strongly Coupled FluidStructure Interaction Problems. 3. Germany 1895. 8. [20] C. M. IDMEC. Sika. Vol. Silva and J. Long Beach. V. Kindervater. Numerical solution of differential-algebraic equations. Simulation of fluid loaded flexible multiple bodies. Pombo and J. Vol.].2000–4823. Nos. [18] H. 13.W. Pombo and J. 6. Vol. Wierzbicki [eds. 1997. 1981. Human Motion Analysis Using Multibody Dynamics ´ o and Optimization Tools. Vol. Lund. Vol. Lisboa. Portugal.4&5. Nonlinear Dynamics. [12] J. 1992. Ambr´sio. [6] J.44–65.S. The static approach to plastic collapse and energy dissipation in some thin-walled steel structures. [15] D. Ambrosio. pp. pp. Technical Report IDMEC/CPM-04/002. 332. [19] N. 2005. Elastic-plastic deformation in multibody dyna´ o mics. Vol. Technical Report IDMEC/CPM-04/001. Instituto Superior T´cnico. Ambr´sio and P.Ambr´sio. 2002.T.]. Kinematic data consistency in the inverse dynamic ´ o analysis of biomechanical systems. Vol. [13] J. Multibody System Dynamics. AIAA Paper No. Glocker.A. 5. 1994. [in:] Crashworthiness Of Transportation Systems: Structural Impact And Occupant Protection. [in:] Structural Crashworthiness.193–207. Continuous contact force models for impact analysis in multibody systems. ´ e [17] G.M. Ambr´sio. Møller. No.525–577.371–407. J. Nikravesh. Hertz Gesammelte Werke . 5. Lund. Multibody Systems Dynamics. [7] F. Tsirakos.

Num.-J. 32. 1983. [23] K.961–986. Int. Test Report. 14.385–400. Anantharaman and M. Large displacement analysis of three-dimensional beam structures.S. and R. Vol. .1531–1542. 16.. Numerical simulation of mechanical systems using methods for differential-algebraic equations.. pp. Int. 1985. [24] M. Num. J. Chung. Bathe and S. Comp. Meth. Struct. Vol.88 ICTAM04 [21] P.L. pp. Methods in Engng. 1979. Vol. Hiller. Eng. The Transportation Research Center of Ohio. 1991. pp. J. Plastic hinge approach to vehicle simulation using a plastic hinge technique. Benedict. [22] 30 mph Rollover Test of an AM General Model M151-A2 1/4 Ton Jeep.E. Nikravesh. Bolourchi. I. J.

Sign up to vote on this title
UsefulNot useful