Kinematic Model of Robots, Direct-Kinematics

(Prepared By: Dr. Bamadev Sahoo, Asst. professor, IIIT, BBSR)

 Kinematic model describes the spatial position of joints and links, and
position and orientation of end-effectors

 The mathematical tools of spatial descriptions developed in the previous

studies [Co-ordinate frames, Mapping, and Transforms] will be used for
modeling of robotic manipulators.

 The kinematic model gives relations between the positions and orientations
of the end effector and spatial positions of the joint-links

 Differential Kinematics of manipulators refer to the differential motion----

Velocity, acceleration and higher order derivatives of position variables

 In this chapter, our immediate objective are

o Link Parameters

o Joint Parameters

o Denavit-Hartenberg Notation (D-H Notation):

o Algorithm to assign the frames to each joint

o Kinematic Modeling of Manipulater

o Applying the Kinematic Model to specific Robotic Manipulators

o Velocity/acceleration, Force/torque associated with motion

Mechanical Structure and Notations

 A manipulator consists of a chain of rigid bodies (called Links), connected to each other
by joints, which allow linear or revolute (rotational/angular) motion between connected
links each of which exhibits just one degree of freedom (DOF).
 Joints with more than one DOF are not common
 A joint with m DOF can be modeled as m joints with 1 DOF each connected with (m-1)
links of zero length.
 Most industrial robotic manipulators are open serial kinematic chains, that is, each link
is connected to two other links.
 The number of DOF a manipulator possesses is the number of independent
parameters required to completely specify its position and orientation in space.
 If each joint has one DOF, then DOF of a Manipulator = No of joints
 1 DOF Joint: (1) Revolute joint(R-Joint) (2) Prismatic joint(P-Joint)

 JOINT AXIS: The axis of relative motion. Here Z-axis in the coordinate frame is
the joint axis.
Outward Numbering of Links: Base=Link 0, first moving body=Link 1, next moving body=Link
2………………Tool/End-effector=Link n

 Description of an object in space requires 6 parameters---3 for position and 3 for

orientation---To position and orient and end effector in 3-D space, a minimum of
3DOF are required for positioning and 3DOF for orientation.
 A typical Robotic Manipulator may have 5 or 6 DOF. This manipulator is assumed
to be consisting of an arm with typically first three links from the base and a
WRIST with 2/3 links.
 The arm accomplishes the task of reaching the desired position, where as the
wrist helps to orient the end-effector.
Descriptions of Links and Joints:
 To describe the position and orientation of a link in space, a frame is assigned to each
link, for example: frame {i} to Link i.
 The position and orientation of frame {i} wrt the previous frame {i-1} is described by the
use of homogeneous transformation matrix.
 Here, we will focus of PARAMETERS required to specify the positions and orientations of
links and joints in a manipulator—LINK PARAMETERS---(1) Link Length(ai) and (2) Link

Twist (αi)

(1) Link Length(ai): The shortest distance along the common normal to the axes i and (i-1)
(2) Link Twist (αi): The angle between the projection of axis (i-1) and axis i, on a plane

perpendicular to the common normal AB. This link twist ( αi) is measured from axis (i-1)

to axis i in the right hand sense about AB as shown in above Fig.3.

(3) For a given link, these two parameters are fixed.
(4) For industrial robots, links are usually straight, the axes (i-1) and i, are parallel, link
length is equal to physical link dimension and link twist equal to zero.
(5) There is another common link geometry exists which is straight link with link twist angle
as multiple of π/2 radians (shown below in Fig.4)
(6) Sometimes, the link may have a bend such that the axis of joint (i-1) and joint i intersect
and in this case the link length of link i is zero although the physical link dimension is not
zero.
 Joint Parameters: (1) Joint distance (di), (2) Joint angle (ϴi)

1. Joint distance (di): Perpendicular distance between two adjacent common normals
ai-1 and ai measured along (i-1). In other words, joint distance is the translation needed
along joint axix (i-1) to make ai-1 intersect with ai

2. Joint angle (ϴi): Angle between the two adjacent common normals a i-1 and
ai, measured in right-handed direction about the axis (i-1). It is the rotation
about joint axis (i-1) needed to make ai-1 parallel to ai
Note: A 1 DOF joint requires only one variable to describe its position. Hence for
every 1-DOF joint, it will always be the case that one the two joint parameters (di
and ϴi) is fixed and the other is variable.
Say qi=Generalized variable representing joint displacement of either type (revolute or
prismatic), then

i ,if joint i is revolute

qi   (1)
di ,if joint i is prismatic
Kinematic Model of a Manipulator: This is the analytical description of the spatial
geometry of motion of the manipulator with respect to a fixed (inertial) reference frame,
as a function of time. In particular the relation between the joint-variables and the position
and orientation of the end-effector is the kinematic model.
It is required to control position and orientation of the end-effector, in 3-D
space, so that it can follow a defined trajectory or manipulate object in workspace.

NOTE: For kinematic modeling, frames are assigned to each link of the manipulator starting
from the base to the end-effector. The homogeneous transformation matrices relating the
frames attached to successive links describe the spatial relationship between adjacent
links. The composition of these individual transform matrices determines the overall
transform matrix, describing tool-frame wrt base-frame.
 Denavit-Hartenberg Notation (D-H Notation): The definition of a manipulator with four
joint-link parameters for each link and a systematic procedure for assigning right-handed
orthogonal coordinate frames, one to each link in an open kinematic chain, was proposed
by Devavit and Hartenberg 1955. This is known as DH Notation.
An n-DOF manipulator will have (n+1) frames---frame {0} is the base frame
(reference inertial frame) and frame {n} is the tool frame.

In the above fig., there are two links i and (i-1); their associated joints (i-1), i and (i+1); and
the axes (i-2), (i-1) and i respectively.
A-B: common normal to axes (i-2) and (i-1),
C-D: common normal to axes (i-1) and i
A frame {i} is assigned to link i as follows
1) The Zi-axis is aligned with axis i, its direction being arbitrary. The choice of direction
defines the positive sense of joint variable ϴi.
2) The Xi-axis is perpendicular to the axis Zi-1 and Zi and points away from axis Zi-1, that is, Xi –
axis is directed along the common normal CD
3) The origin of the ith coordinate frame, frame {i}, is located at the intersection of axis of
joint (i+1), that is, axis i, and the common normal between axes (i-1) and i( here, common
normal is CD).
 4) Finally, Yi-axis completes the right-hand orthogonal coordinate frame {i}. It may be noted
that the frame {i} for link i is at the distal end of link i and moves with the link.

 With respect to frame {i-1} and frame {i}, the four DH-parameters --- Two link
parameters (link length(ai ) and link twist(αi)) and two joint parameters ((1) Joint distance (di),
(2) Joint angle (ϴi))----are defined as
1) link length(ai): Distance measured along Xi- axis from the point of intersection of Xi-
axis with Zi-1 –axis (point C) to the origin of frame {i}, that is, distance CD.
2) link twist(αi):-Angle between Zi-1 - and Zi –axes measured about Xi axis in the right
handed sense.
3) Joint distance (di): Distance measured along Zi-1 – axis from origin of frame (i-1) (point B)
to the intersection of Xi–axis with the Zi-1–axis (point C), that is distance BC.
4) Joint angle (ϴi): Angle between Xi-1 - and Xi –axes measured about Zi-1 axis in the right
handed sense.
NOTE: The conventions outlined above do not result in a unique attachment of frames to links because
alternative choices are available. For example, joint axis i has two choices of direction to point Zi –axes,
one pointing upward and other pointing downward. To minimize such options and get a consistent set
of frames, an algorithm is presented to assign frames to all links of a manipulator.
 Algorithm for Link-Frame Assignment:
 This algorithm assigns frames and determines the DH-parameters for each link of
an n-DOF manipulator.
 First link (Frame {0}) and the last frame (Frame {n}) are assigned after assigning
frames to the intermediate links, i.e. link 1 to link (n-1).
 Home position of the manipulator must be defined. qi=0 for i=1, 2, 3, 4………n
 The algorithm has 4-parts:
I. Steps for labeling scheme

II. Steps for frame assignment to intermediate links 1 to (n-1)

III. Steps to assign frame {0} to the base
IV. Steps to assign frame {n} to the Tool/end-effector
I. Steps for labeling scheme
Step 0: Identify and number the joints from starting with base(0) and ending with end-
effector(n).
Step 1: Align axis Zi with axis of joint (i+1) for i=0, 1, ….. (n-1)
II. Steps for frame assignment to intermediate links 1 to (n-1)
For each link i repeat steps 2 and 3
Step 2: The Xi-axis is defined perpendicular to both Zi-1 - and Zi –axes and points away from Zi-

The origin of frame {i} is located at the intersection of Zi – and Xi – axes. Three situations are
possible:
Case (i): If Zi-1 - and Zi –axes intersect, choose the origin at the point of their intersection.
The Xi –axis will be perpendicular to the plane containing Zi-1 - and Zi –axes. This will give

ai=0.
Case (ii): If Zi-1 - and Zi –axes are parallel or lie in parallel planes then their common normal
is not uniquely defined. If joint i is revolute then Xi –axis is chosen along the common
normal, which passes through origin of frame {i-1}. This will fix the origin and make di=0. If
joint i is prismatic, Xi –axis is arbitrarily chosen as any convenient common normal and the
origin is located at the distal end of the link i.
 Case(iii): If Zi-1 - and Zi –axes are coincident, the origin lies on the common axis. If joint i is
revolute, origin is located to coincide with origin of frame {i-1} and Xi –axis coincides with
Xi -1–axis to cause di=0. If joint i is prismatic, Xi –axis is chosen parallel to Xi -1–axis, to make

ai=0. The origin is located at the distal end of link i.

Step 3: The Yi-axis has no choice and is fixed to complete the right-handed orthogonal
coordinate frame {i}.
III. Steps to assign frame {0}: To the immobile base

Step 4: The choice is made based on simplification of model and some convenient
reference in workspace. The X0–axis, which is perpendicular to Z0–axis, is chosen to be parallel
to X1–axis in the home position to make ϴ1=0.
The origin of frame {0} is located based on type of joint 1. If the joint 1 is revolute, the
origin of frame {0} can be chosen at a convenient reference such as floor, work table and so
on, giving a constant value of parameter d 1 or to make d1=0.
If the joint 1 is prismatic, parallel X0– and X1–axes will make ϴ1=0, and the origin of
frame {0} is placed arbitrarily.

Step 5: The Y0-axis completes the right handed orthogonal coordinate frame {0}.
IV. Steps to assign frame {n}: To the end-effector

Step 6: The origin of frame {n} is chosen at the tip of the manipulator, that is, a
convenient point on the last link (the end-effector). This point is called the “tool point”
and the frame {n} is the tool frame.
Step 7: The Zn-axis is fixed along the direction of Zn-1 axis and pointing away from the link
n. It is the direction of approach.
Step 8: If joint n is prismatic, take Xn parallel to Xn-1 –axis. If joint n is revolute, the choice
of Xn is similar to step 4; that is, Xn is perpendicular to both Zn-1 and Zn axes. Xn direction
is the “normal” direction. The Yn-axis is the “orientation” or the “sliding direction.
Note: Once the frames are assigned to each link, the joint-link parameters (ϴi , di. αi, ai)
can be easily identified for each link, using which the direct kinematic model is
developed. In fixing frames it is desirable to make as many of the joint-link parameters
 zero as possible because it will reduce the amount of computation in the subsequent
steps.
Kinematic Relationship between Adjacent Links:
(Aim is to find the Link i Transformation Matrix: i 1Ti ( qi ) )
 Consider frame {i-1} and frame {i} are associated with link (i-1) and i respectively (links
not shown).
 Kinematic Joint-Link parameters: ϴi , di , αi , ai
 In Fig. 9 points B, C, D are same as in Fig.8: Recall, AB and CD are common normal

 Transformation of frame {i-1} to frame {i} consists of 4 basic transformations as shown

in Fig. 9. They are
 A rotation about Zi-1 –axis by an angle ϴi
 Translation along Zi-1 –axis by distance di
 Translation along Xi –axis by distance ai
 Rotation about Xi –axis by an angle αi
 Using the spatial coordinate transformation matrix, the composite transformation matrix,
which describes frame {i} wrt frame {i-1} can be obtained like
Ti  Tz (i )Tz (di )Tx (ai )Tx (i )
i 1
(2)
 Ci  Si 0 0 1 0 0 0  1 0 0 ai  1 0 0 0 
 S  0   
Ci 0 0 1 0 0  0 1 0 0  0 C i  S i 0 
i 1
Ti   i 
0 0 1 0 0 0 1 di  0 0 1 0  0 S i C i 0 
     
0 0 0 1  0 0 0 1  0 0 0 1  0 0 0 1 
Ci  Si C i Si S i ai Ci 
 S 
Ci C i  Ci S i ai S i 
 i (3)
0 S i C i di 
 
0 0 0 1 

Alternate Approach to Find i 1Ti : (Transformation from frame {i-1} to frame {i})

 Introduce an intermediate frame {i’} at point C between frame {i-1} and frame {i}
 Transformation from frame {i} to frame {i’}: Two Operation
I. Rotation about Xi axis:  i

II. Translation about Xi axis: ai

1 0 0 ai 
0 
C i  S i 0
i'
Ti   (4a)
0 S i C i 0
 
0 0 0 1 

 Transformation from frame {i’} to frame {i-1}: Two Operation

I. Rotation about Zi-1 axis:  i

II. Translation about Zi-1 axis: d i

C i  S i 0 0
 S 
C i 0 0
i 1
Ti '   i (4b)
0 0 1 di 
 
0 0 0 1 
 The composite transformation from frame {i} to frame {i-1}
i 1 i 1
Ti  Ti ' i 'Ti
Ci  Si 0 0  1 0 0 ai 
 S  
Ci 0 0  0 C i  S i 0
 i
0 0 1 di  0 S i C i 0
  
0 0 0 1  0 0 0 1 
Ci  Si C i Si S i ai Ci 
 S 
Ci C i  Ci S i ai Si 
 i (5)
0 S i C i di 
 
0 0 0 1 

Important Note:
 Equation (3) and (5) are identical
i 1
 The homogeneous transformation matrix Ti describes the position and orientation
of frame {i} relative to frame {i-1} and completely specifies the geometric relation
between these links in terms of four DH-parameters (i , di ,  i , ai ) .

 Of these four parameters, only one is a variable for link i, the displacement
(generalized) variable qi (i or di ) and the other three are constant.
i 1
 The matrix Ti ( qi ) is known as link i transformation matrix.
 3x3 upper left corner sub-matrix: Gives orientation of coordinate axes of frame {i}.
 3x1 upper right corner sub-matrix represents the position of origin of frame {i}.
Manipulator Transformation Matrix:

Recall, the previous concept,

Frame {0}-----Base(Link 0)
Frame{1}------Link1
………………………..
……………………………
Frame {n}------Tool point/end-effector
Total no of frames={n+1}, Total no links=(n+1), Total no of joints=n

i 1
Ti ( qi ) : Homogeneous Link Transformation Matrix between frames {i-1} and {i}
or
Transformation Matrix of Link i
qi : Generalized joint displacement variable (i or di )
i  1, 2,3,................n
The position and orientation of tool frame relative to base frame can be
found by considering the n consecutive link transformation matrices relating
frames to adjacent links

0
Tn  0T1 (q1 ) 1T2 (q2 ) 2T3 (q3 )........ n1Tn (qn ) (6)
Alternate Approach: (Assigning (n o a) frame/coordinate system to an end-
effector))

Recall the End-effector,

Z-axis=Approach vector (direction in which end-effector approaches towards the target)

Y-axis=Sliding/Orientation vector (direction of opening and closing the end-effector as it
manipulates the object.
X-axis=Normal vector to approach and sliding vectors.

T=End-effector Transformation Matrix in terms of tool frame orientation (n, o, a) and the
displacement (d) from the base frame {0}

In Fig.10, frame {n} is the tool frame, thus T  0Tn

T  0Tn  0T1 1T2 .............. n1Tn (7)

Important Note:
 Eq.(7) is the Kinematic Model of the n-DOF Manipulator.
 It provides the functional relationship between the tool frame (or, end-effector)
position and orientation, and displacement of each link qi , which may be angular
or linear depending on the joint being revolute or prismatic.
 T  f (qi ), i  1, 2,3................n (8)

 nx ox a x d x   r11 r12 r13 r14 

n  
oy a y d y   r21 r22 r23 r24 
Or,  
y
(9)
 nz oz a z dz  r r32 r33 r34 
   31 
0 0 0 1  0 0 0 1 
Where rij  f ( qi )
 For the known joint displacements qi , i  1, 2,3............n , the end-effector
orientation (n o a) and position d can be computed from Eq.(8).