Download full document - Automatic Control
Download full document - Automatic Control
Download full document - Automatic Control
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
5 TRAJECTORY<br />
A trajectory is a path that an object follows through space. In robotics the object could be the<br />
end-effector for a serial robot or the TCP for a Delta-3 robot. A trajectory can be described<br />
mathematically either by the geometry of the path, or as the position of the object over time.<br />
There are some parameters that have to be decided when a trajectory will be calculated. For<br />
example if the motion should be linear, quadratic or cubic, also if the motion should be done<br />
with some predefined start and end velocity, constant velocity or for example with minimum<br />
time. Then with this information a trajectory is calculated with help of interpolation. This<br />
trajectory contains then information about the speed, acceleration and position.<br />
5.1 TRAJECTORY WITH DESIRED START AND END VELOCITY<br />
One way to generate a smooth trajectory for an object, for example a time history of desired<br />
end-effector coordinates is by a polynomial (5). In this case there are some constraints such as<br />
the desired start and end velocities. If the variable X is a vector that describes the coordinates<br />
(x,y,z) for the end-effectors position. Then suppose that the object coordinates at time t 0<br />
satisfies<br />
and the final values for time t f satisfies<br />
X ( t0<br />
) = X<br />
X&<br />
( t ) = X&<br />
0<br />
Eq. 5.1<br />
f<br />
X&<br />
( t ) = X&<br />
f<br />
Eq. 5.2<br />
______________________________________________________________________________<br />
Public Report ELAU GmbH, Marktheidenfeld<br />
28<br />
0<br />
0<br />
X ( t ) = X<br />
Because there is four constraints the generated polynomial has to contain four independent<br />
coefficients that can be chosen to satisfy these four constraints. This will give a cubic trajectory<br />
of the form<br />
X d<br />
( t)<br />
= a + a t + a t + a t<br />
0 1<br />
Eq. 5.3<br />
where Xd (t)<br />
is the desired (x,y,z) positions at time t.<br />
Then the generated velocity that will satisfy the start and end constraints will be given by the<br />
derivative of Xdesired as<br />
f<br />
f<br />
2<br />
2<br />
3<br />
3