The Optimal Design of Three Degree-of-Freedom Parallel Mechanisms For Machining Applications
The Optimal Design of Three Degree-of-Freedom Parallel Mechanisms For Machining Applications
The Optimal Design of Three Degree-of-Freedom Parallel Mechanisms For Machining Applications
Introduction
2
2.1
P
Figure 2a: A serial
singularity
Singularities
The determination of the singular configurations for twodof mechanisms is very simple; conversely, for a six-dof
mechanism like the Gough-Stewart platform, a mechanism
with six-dof, the problem is very difficult [3]. With a
modular architecture, when the position and the orientation
of the mobile platform are decoupled, the determination of
the singularities is easier.
2.3
i min
i max
A1
e2
P
j
x
z
velocity
of point
P can be
two
expressed in
&
where a1 , b1 , a2 and b2
represent the position vectors
of the points A1 , B1 , A2 and B2 , respectively. Moreover,
of A1 and A2
the velocities & a and &
are given by
a
&
&
&
&
and
a e
2 e , respectively.
For an isotropic configuration to exist where the velocity
amplification factors are equal to one, we must have e1.e2
0 [11] (Figure 5).
3.1
of
parallel
mechanism
for
2 i (b2 a2 )
(1b)
&
1 1
2 2
&
translational motions
The aim of this section is to define the kinematics and the
singular configuration of a two-dof translational mechanism
(Figure 3), which can be extended to three-axis machines
by adding a third axis in series with the first two. The
output body is connected to the linear joints
through a set of two parallelograms of equal lengths
L Ai Bi , so that it can move only in translation.
The two legs are PPa identical chains, where P and Pa
stand for Prismatic and Parallelogram joints, respectively.
This mechanism can be optimized to have a workspace
whose shape is close to a square workspace and the velocity
amplification factors are bounded [11].
The joint variables 1 and 2 are associated with the two
prismatic joints. The output variables are the Cartesian
coordinates of the tool center point P [x y]T . To control
the orientation of the reference frame attached to P, two
parallelograms can be used, which also increase the rigidity
of the structure, Figure 3.
(2a)
T
a )T e
(b
A
B
1
1
T
where (b2 a2 )
&
and p is the velocity of point P, i.e.,
&
0
1
(b2 a2 )T e2
&
&
&
& &T
&
and p x
y
When A and B are not singular, we obtain the relations,
&
&
p J with J A
A1 B1 and
Parallel singularities occur whenever the lines
1
A2 B2 are colinear, i.e. when 2
k , for k = 1,2,....
3.2
B1
B2
p a2
2 i ( b2 a2 )
(3b)
p a3
( 3 j k)
& &
a1 e11 ,
and p [x y ]
a3 e3 3 , respectively.
&
and
&
(b3 a3 ) p
&
&
(b3 a3 ) e3 3
(4c)
&
(b3 a3 ) j (p b3 )
& &&T
t J with J A
t [x y ]
B and
0
T
) j (p
a b
(b a ) T e
x
e22
(b a )T (b
a2
A (b2
(3c)
a2 )
&
( b3 a3 ) j (p b3 )
&
[1
&
e2
&
A2
e1
(3a)
& &
&
1 i (b1 a1 )
A1
& &
p a1
0
(b2 a2 )T e2
0
0
0
0
a )T e
(b
3
first one
)T j (p
(b 3 a 3 b
is
3
parallel
singularity
when
( A B ) and
3
e1
e2
e1
A1
e2
A1
B1
P
(a)
B2
B1
Cartesian workspace
Useful workspace
Close to singularity locus
B2
(b)
P1
P2
P2
x
(a)
motions
P3
and e3
e3 L1
(b)
A3
(a)
e3
B3
A
(b)
e3
3
p b3
(Figure 7a).
x
P1
3.4
P A3
(c)
A3
e3
3
(d)
obtained for the solution (b) is smaller than for the solution
(a) but is more appropriate for the extension three-axis
mechanism of Figure 4. In effect, we want the axis of
rotation to be parallel to one of the side of the
configuration
whenever
(Figure 7b-c) or (b3 a3 )T e3
optimization function, we set:
(b3 a3 )T j (p b3 )
0
0 (Figure 7d). In the
b3 a3
)T
j
(p
b 3)
p b3
0,2
(4a)
(4b)
A3 B3 e3
is in
3 3
is
78,5 78,5 .
optimization as a function of
Fig. 8.
(b3 a3 )T
b3 a3 e3 0,2
(b 3 a
in 11,5168,5
and
L1 and L2 is depicted in
L2=1,8L1
2 4.0
3.5
2.5
2.0
L
1.5
2.0
2.5
Montreal, 2001.
1
3.0
[6] Gao, F., Li, W., Zhao, X., Jin Z. and Zhao. H., New
kinematic structures for 2-, 3-, 4-, and 5-DOF parallel
optimal.
Conclusions
References
[1] Treib, T. and Zirn, O., Similarity laws of serial and parallel
manipulators for machine tools, Proc. Int. Seminar on
Improving Machine Tool Performances, pp. 125-131, Vol. 1,
1998.