6 Robot Kinematics and Dynamics 1 PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 45

6.

Basic Coordinates
Rotation and Translation in basic of Robot Coordinates, Homogeneous
g
Transform
Euler Angles, Denavit-Hartenberg Notation

Affiliated Professor : Jaeyoung Lee


E-mail : [email protected]
H P : 011 9455 - 5277
H.P

IT

Topics
p

Basic Coordinates

Coordinate Frames & Denavit-Hartenberg


Denavit Hartenberg Notation

Forward and Inverse Kinematics of Robot Manipulators

Manipulator
p
Jacobian

Review of Lagrangian Dynamics

Lagrangian Dynamics of Robot Manipulators

Newton-Euler Equations of Robot Manipuators

Forward and Inverse Dynamics

Robot Statics

Robot Trajectory Planning

Robot Control

IT

Basic Coordinates

Operators for Translation and Rotation

Translation

Rotation

IT

Basic Coordinate

Coordinate Frames

Translation

Rotation

IT

Basic Coordinate

Different View Based Upon Different Coordinates

Translation

Rotation

IT

Basic Coordinate

Rotation and Translation Combined

A single rotation after a single translation:

A single rotation and a single translation after another pair of


translation and rotation:

Scheme shown above to represent transform operations is not a


good way of simplifying notation and computations.

We need consistent transforms for translations and rotations:


homogeneous transform

IT

Basic Coordinate

Rotation Matrix R

Vector p can be represented with respect to the


two different coordinates, i.e., coordinate A and B

We can to find a rotation matrix such that

Noting that x_A, y_A and z_A represent the


components of p along the x, y, and z axes of
coordinate A.

IT

Basic Coordinate

Rotation Matrix R

Noting that x_A,


x A, y_A
y A and z_A
z A represent the components of p along the x, y, and z
axes of coordinate A.

Therefore,

IT

Basic Coordinate

Rotation Matrix R

Therefore,

Similarly,

Thus,

Rotational matrix R is orthonormal (orthogonal).

Therefore,

IT

Basic Coordinate

Homogeneous Transform

In homogeneous transforms, translation and orientations have identical


characteristics.

Position vector [x y z]T is represented by an augmented vector [x y z 1]T .

Translation is represented by a matrix:

10

IT

Basic Coordinate
-

Rotation is represented by a matrix:

11

IT

Basic Coordinate
-

If translation and rotation are combined

12

IT

Basic Coordinate
-

If translation and rotation are combined


Therefore,

Inverse of homogeneous transform

13

IT

Basic Coordinate

More On Rotation : Euler Angles

Many different sets of Euler angles exist

Z-Y-Z, Z-Y-X, and etc,

Example : Z-Y-X : Rotate around z-axis, and then around y-axis, and finally
around x-axis
x axis.

Rotational Transformation (Post-multiplication)

14

IT

Basic Coordinate

More On Rotation : Euler Angles

15

IT

Basic Coordinate

More On Rotation : Euler Angles


X-Y-Z
X
Y Z Fixed Angles (Pitch-Yaw-Roll)
(Pitch Yaw Roll)

Note

Rotational Transformation (Pre


(Pre-multiplicatin)
multiplicatin)

16

IT

Basic Coordinate

Equivalent Angle-Axis

Represented by a unit vector that indicates the axis of rotation and an angle to
rotate.

If the axis of rotation is k = [K_x K_y K_z]^T,

17

IT

Basic Coordinate

Denavit-Hartenberg Notation

Conventions

Joint i moves link i


Frame i is fixed at (attached to) link i and located at the distal joint, which means
that frame i is located at joint i+1.
Z-axes are aligned with joint axes.

18

IT

Basic Coordinate

4 Link Parameters

Link Length(L_n):
Length(L n): the shortest distance between two joint axes at both
ends of Link n.

Link Twist(alpha_n): "twist" angle of two z-axes at both ends of Link n


in the direction of x_n
x n -axis
axis.

Link Offset (d_n): distance from the origins of Frame n-1 to Frame n in
the direction of Z_n-1-axis.

- Joint
i Angle(Thetat_n):
l h
angle
l ffrom x_n-1-axis
1 i to x_n -axis
i iin the
h direction
di
i
of z_n-1-axis.

19

IT

Basic Coordinate

Miscellaneous Conventions

If z-axes
z axes meet each other,
other x_n
x n axis,
axis in general,
general is defined to be zz_n
n1z_n.

Positive direction of z-axis can be either.

20

IT

Basic Coordinate

Miscellaneous Conventions

The origin of frame 0 is usually coincide with that of frame 1.


1

Z-axis of frame 0 is aligned with z-axis of joint 1.

For a prismatic joint, z-axis is aligned with the sliding direction of the
j i IIn this
joint.
hi case, link
li k offset
ff
d iis the
h jjoint
i variable.
i bl

Homogeneous transform :

Homogeneous transform can be obtained directly by observing the


coordinates
d
in most cases.

21

IT

Basic Coordinate

Example :

22

IT

Basic Coordinate

Example :

Note that
h T^0_1 can be
b directly
d
l determined
d
d

23

IT

Kinematics & Inverse Kinematics

Forward Kinematics

For a given set of joint displacements, the end-effector position and


orientation can be calculated. (Forward Kinematics)

Inverse Kinematics

For a given set of end-effector position

and orientation, joint displacements are


computed.
-

We have to solve nonlinear equations;

however, we may not able to solve for


solutions analytically.

24

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Forward Kinematics

25

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


F
Forward
d Kinematics
Ki
ti

26

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


E d Eff t P
End-Effector
Position
iti

End-Effector Orientation

27

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator

Inverse Kinematics
- For a given tip position [x y]^T, the joint displacement [theta_1 theta_2]^T
is to be determined.
Geometric Method
(Using cosine rule)

28

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator

Inverse Kinematics

29

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator

Inverse Kinematics

30

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator

Inverse Kinematics

Sign: "+", if -p < theta_2 < 0, i.e., at "elbow up configuration "-", otherwise.

31

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Algebraic Method

left hand side: a constant matrix


left-hand-side:
right-hand-side: a matrix whose elements are functions of joint angles

32

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Algebraic Method

33

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Algebraic Method

Let
Then

34

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Algebraic Method
Similarly,

Using
g

35

IT

Kinematics & Inverse Kinematics

Kinematics & Inverse Kinematics for a 2-D Revolute-Jointed Manipulator


Algebraic Method
where

Therefore,

36

IT

Kinematics & Inverse Kinematics

Kinematic Decoupling
If the final 3 consecutive axes meet at a point, the process of solving inverse
kinematics is simplified.
Three axes z4 , z5 , and z6
meet at a wrist center
center.

The position of the wrist center, w, is determined


by q1, q2 , and q3 only.

37

IT

Kinematics & Inverse Kinematics

Kinematic Decoupling
- From the target tool position d and orientation R, wrist center
position can be computed:
where
h
- Once the wrist center is known, the first 3 joint variables, q1, q2, and q3 , can
be computed.
- The orientation matrix R^0_3 can be computed after q1, q2 , and q3 are
d t
determined.
i d
- The rest of joint variables, q 4, q5 , and q6 , can be computed from R_6^3,
which in turn can be computed by

38

IT

Kinematics & Inverse Kinematics

SCARA Robot

39

IT

Kinematics & Inverse Kinematics

SCARA Robot

- Inverse Kinematics of SCARA Robot

40

IT

Kinematics & Inverse Kinematics

SCARA Robot

- Rotation matrix R can be in the form of

41

IT

Kinematics & Inverse Kinematics

SCARA Robot

Using a geometry formed by projecting the manipulator configuration


upon a horizontal plane, which forms a 2-link manipulator,

Again, similarly to a 2-link manipulator,

and

42

IT

Kinematics & Inverse Kinematics

SCARA Robot

- Solvability : A manipulator is defined to be solvable, if all the sets of joint variables


for a given position and orientation can be determined by an algorithm. Many
manipulators have multiple solutions for a single configuration. For example,
PUMA 560 has 8 solutions.
-W
Workspace
k
: The
Th volume
l
off space that
h the
h end-effector
d ff
off a manipulator
i l
can reach.
h
Within the workspace, inverse kinematics solutions exist.
- Closed Form Solution : Inverse kinematic solutions can be grouped into closed form
solutions and numerical solutions.
solutions Closed form solutions are noniterative such as
algebraic expressions; whereas numerical solutions are iterative. Numerical
solutions are computationally more expensive and slower.
p
work : A g
general 6 d.o.f. manipulator
p
does not have a closed form solution.
- Pieper's
If the three consecutive axies intersect at a point, a closed form solution exists. For
most commercial manipulators, the last consecutive axes intersect at a point.
- Notes: Even if inverse kinematic solutions are found, they may not be physically
realizable due to joint angle limitations
limitations.

43

IT

HW 6

139p, 1

44

IT

REFERENCE

2 :

Introduction to Robotics,, byy J,, J. Craig


g

Ch 2. Spatial descriptions and transformations

Ch 3. Manipulator kinematics

Ch 4. Inverse manipulator kinematics

45

You might also like