homogenous transformation matrices

19
T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010 HOMOGENOUS TRANSFORMATION MATRICES T. Bajd and M. Mihelj

Upload: amie

Post on 22-Jan-2016

91 views

Category:

Documents


2 download

DESCRIPTION

HOMOGENOUS TRANSFORMATION MATRICES. T. Bajd and M. Mihelj. Homogenous matrix. • • • □ • • • □ • • • □ ■ ■ ■ 1. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

HOMOGENOUSTRANSFORMATION

MATRICES

T. Bajd and M. Mihelj

Page 2: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

The homogenous matrix describes either pose (orientation and position) or displacement (rotation and translation) of an object. It consists of a rotation matrix (•), translation column (□), and perspective transformation row (■).

• • • □ • • • □ • • • □ ■ ■ ■ 1

Homogenous matrix

Page 3: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

The elements of rotation matrix are direction cosines of the angles between individual axes of the coordinate frames and .

Homogenous matrix

Rotation matrix

Page 4: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

The first three rows correspond to the and axis of the reference frame, while first three columns refer to and axis of the rotated frame. The element of the matrix is cosine of the angle between the axes given by the corresponding column and row.

Rotation about axis

Homogenous matrix

Page 5: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

1000

0cos0sin

0010

0sin0cos

),(

yRot

x’ y’ z’

xyz

1000

0cossin0

0sincos0

0001

),(

xRot

x’ y’ z’

xyz

x’ y’ z’

xyz

Homogenous matrix

Rotation about and axes

Page 6: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

1000

100

0010

0001

),(c

czTrans

xyz

1000

0100

010

0001

),(b

byTrans

xyz

1000

0100

0010

001

),(

a

axTrans

xyz

Homogenous matrix

Translation along and axes

Page 7: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Pose of frame with respect to reference frame

From the homogenous matrix we „read“, that the axis has the same direction as axis of the reference frame, axis the same direction as axis, while axis is directed in the same way as axis.

Pose

Page 8: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Displacement of a frame with respect to a relative coordinate frame

The homogenous matrix H can be explained by three successive displacements of the reference frame.

Displacement

Page 9: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Position and orientation of the first block O1 with respect to the base block O0.

Homogenous matrix

Page 10: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Homogenous matrix

Position and orientation of the second block O2 with respect to the first block O1.

Page 11: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Homogenous matrix

Position and orientation of the third block O3 with respect to the second block O2.

Page 12: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Position and orientation of the third block O3 with respect to the base block O0 is obtained by successive multiplications of the three matrices.

Homogenous matrix

Page 13: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

The correctness of the calculated orientation and position of the third block O3 with respect to the base block O0 can be easily verified from the figure.

Homogenous matrix

Page 14: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

When second block rotates around axis 1, and the third block around axis 2, while the last block is elongated along axis 3, the so called SCARA robot is obtained.

Geometric robot model

Page 15: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Because of displacements about axis 1 and 2 and along axis 3, the homogenous matrices consist of products of first matrix describing the pose of the object and second matrix describing its displacement.

Geometric robot model

Page 16: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Geometric robot model

Pose and rotationin the first joint

Page 17: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Pose and rotationin the second joint

Geometric robot model

Page 18: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

Geometric robot model

Pose and rotationin the third joint

Page 19: HOMOGENOUS TRANSFORMATION  MATRICES

T. Bajd, M. Mihelj, J. Lenarčič, A. Stanovnik, M. Munih, Robotics, Springer, 2010

The geometric model of a robot describes the pose of the frame attached to the end-effector with respect to the reference frame on the robot base. It is obtained by successive multiplications of homogenous matrices.

Geometric robot model