q
= [ q
1
, q
2
... q
n
]
T
∈ R
n
(5.5)
If the number of robot drives is equal to or exceeds six drives, then the end-e
ffector is
capable of a six DOF motion in spatial space. At any moment in time, the posture of the
end-e
ffector w.r.t the robot’s frame can be expressed as a function of the joint variables
x
R
= f (q)
(5.6)
To obtain expressions for the velocity and acceleration of the TCP, equation (5.6) is
di
fferentiated twice w.r.t time, yielding the following expressions
˙x
R
= ˙f(q) = J(q)˙q
(5.7)
¨x
R
= ˙J(q)˙q + J(q)¨q
(5.8)
where J(q)
∈ R
6
×n
is termed the jacobian matrix and is defined as
J(q)
=
∂x
∂q
(5.9)
Depending on the manipulator configuration, the jacobian matrix represents mapping
from the joint space to the task space of the TCP. It constitutes one of the most useful
mathematical tools in robotics. It is utilized to find singular configurations of a manipulator,
to analyze redundancy and even to map the joint torques to the forces
/torques arising on
the TCP (Sciavicco & Siciliano 2005).
59
5 Control Architecture
5.2.3.2 Analysis for cooperative manipulators
In order to extend the latter definitions to a group of cooperative manipulators interacting
with the same work-piece, the following is assumed:
• Assumption 1: The system exhibits flexibility only through the control system and
not due to any flexibility in the work-piece, hence the manipulators and the object
are considered rigid and non-deformable (Kosuge et al. 1997)(Reinhart & Zaidan
2009).
• Assumption 2: The grasp between the manipulators and the object at the gripping
point is tight i.e. no slipping or sliding occurs (Caccavale et al. 2000).
• Assumption 3: The jacobian matrices either in an analytical or in a geometric form
are known and hence the kinematics of the manipulators are known (Gueaieb et al.
2007a).
• Assumption 4: The geometrical and dynamical properties of the object are known
(Gudi˜no Lau & Arteaga 2005).
The starting point of this analysis are the 1
st
and 2
nd
assumptions. According to which the
distance between the object frame and any tool frame is constant throughout the movement
of both the manipulators and the object. Consequently, the rotation of the TCP frame of
any manipulator i is equal to the rotation of the object frame o
θ
i
= θ
o
(5.10)
or in rotation matrix form:
R(
θ)
i
= R(θ)
o
(5.11)
according to Figure 5.3, vector
r
(
O−T
i
)
defines the distance between the the two frames
O
and
T
i
, which is constant throughout the motion. Hence, the position of the tool frame
T
i
w.r.t. the world frame is equal to
r
i
= r
o
+ R(θ)
o
r
(
O−T
i
)
= r
o
+ r
θ
o
(5.12)
where
r
θ
o
= R(θ)
o
r
(
O−T
i
)
(5.13)
60
5.2 Modeling
W
O
‘‘
T
i
‘‘
O
‘
T
i
‘
Do'stlaringiz bilan baham: |