ROBOT HAND POSTURE TASK SPACE ANALYSIS

BY COMPOSITE HOMOGENOUS
TRANSFORMATION MATRICES
Madanapuri Naveen1

Mr. K. Rajesh Babu2

1Student, Department of Production Engineering, Sri Venkateswara University College of

Engineering, Tirupati-517502, (India)

2
Associate Professor, Department of Production Engineering, Sri Venkateswara University College of
Engineering, Tirupati-517502, (India)

1
Corresponding Author: Madanapuri Naveen [email protected]

ABSTRACT: Applications of robots in various fields like, industrial automation, mining, space application,
medical science, agriculture application etc. have been increased with recent need and technology development.
The objective of this project is Forward kinematic analysis of Scara robot. Using the Scara robot, the forward
kinematic analysis has been done from home position to desired position. End effector without offset distance and
with offset distance to the robot arm. Euler angles, roll, pitch and yaw and final position of robot arm from home
position to desired position has been determined. Therefore, improve the accuracy, reduce the human effort and
save the time. The outcomes of this project have significant implications for the agricultural and industries areas
like plucking of fruits, pick and place of objects in industries and also pave the way for further advancements in
robotics.
Keywords: Forward kinematic analysis, Robotics, Scare robot, Euler angles and Roll, Pitch and Yaw.

1. INTRODUCTION

A SCARA (Selective Compliance Assembly Robot Arm) robot's forward kinematic analysis entails figuring out
the end-effector's position and orientation based on the robot's joint angles. An industrial robot that is widely
utilized for pick-and-place and assembly tasks is the SCARA robot. A SCARA robot's forward kinematics are
often computed using a geometric method. In order to do this, the connections and joints of the robot must be
represented as a set of rigid transformations in a coordinate system. It is possible to determine the end-effector's
position and orientation.

The Denavit-Hartenberg (DH) parameters are frequently used to represent the kinematic chain in forward
kinematics. These variables offer a methodical manner to explain the geometric arrangement of nearby linkages
and joints. The transformation matrices that represent the link transformations can be derived using the DH
parameters. The transformation matrix from the base frame to the end-effector frame can be obtained by
multiplying the transformation matrices for each joint together. Then, using this transformation matrix, the
position and orientation of the end-effector may be determined and Euler angles and roll, pitch and yaw can be
determined with out set distance of end effector and with offset distance of end effector.
 2. METHODOLOGY
2.1 Forward kinematics of 4DOF of Scara robot

Fig.2.1 Forward Kinematic solution for a 4 DOF Scara robot arm

The homogeneous transformation matrix T describing the position and orientation of the coordinate
frame of end of the arm with respect to coordinate frame of base frameis written as

0 0 1 2 3
T4 = T1 * T2* T3* T4

In order to calculate the final transformation matrix T, DH parameters for each in must bedefined.

Table 2.1: DH Parameters for 4-DOF Scara robot arm

Link 𝜽𝒊 𝒅𝒊 𝜶𝒊 𝒂𝒊
1 𝜃1 𝑑1 180 a1

2 𝜃2 0 0 a2

3 0 𝑑3 0 0

4 𝜃4 𝑑4 0 0

0 0 1 2 3
T4 = T1 * T2* T3* T4

cos⁡(𝜃1 ) −sin⁡(𝜃1 ) 0 𝑎1 ∗ cos⁡(𝜃1 ) cos⁡(𝜃2 ) −sin⁡(𝜃2 ) 0 𝑎2 ∗ cos⁡(𝜃2 )

0 sin⁡(𝜃1 ) ⁡cos⁡(𝜃1 ) 0 𝑎1 ∗ sin⁡(𝜃1 ) sin⁡(𝜃2 ) ⁡cos⁡(𝜃2 ) 0 𝑎2 ∗ sin⁡(𝜃2 )
T4=[ ]*[ ]*
0 0 −1 𝑑1 0 0 1 0
0 0 0 1 0 0 0 1
1 0 0 0 cos⁡(𝜃4 ) −sin⁡(𝜃4 ) 0 0
0 ⁡1 0 0 sin⁡(𝜃4 ) ⁡cos⁡(𝜃4 ) 0 0
⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡[ ]*[ ]
0 0 ⁡1 𝑑3 0 0 ⁡1 𝑑4
0 0 0 1 0 0 0 1
 cos⁡(𝜃1 − 𝜃2 − 𝜃4 ) sin⁡(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑎1 ∗ cos(𝜃1 ) + 𝑎2 ∗ cos⁡(𝜃1 − 𝜃2 )
0
T4=[ sin⁡(𝜃1 − 𝜃2 − 𝜃4 ) −cos⁡(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑎1 ∗ sin(𝜃1 ) + 𝑎2 ∗ sin⁡(𝜃1 − 𝜃2 ) ]
0 0 −1 𝑑1 − 𝑑3 − 𝑑4
0 0 0 1

Let us take x=𝑎1 ∗ cos(𝜃1 ) + 𝑎2 ∗ cos⁡(𝜃1 − 𝜃2 )

y=𝑎1 ∗ sin(𝜃1 ) + 𝑎2 ∗ sin⁡(𝜃1 − 𝜃2 )

z=𝑑1 − 𝑑3 − 𝑑4
cos(𝜃1 − 𝜃2 − 𝜃4 ) sin(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑥 R11 R12 R13 𝑥
0
T4=[ sin(𝜃1 − 𝜃2 − 𝜃4
) − cos(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑦]=[R21 𝑅22 𝑅23 𝑦
]⁡
0 0 −1 𝑧 𝑅31 𝑅32 R33 𝑧
0 0 0 1 0 0 0 1
2.2 Forward kinematics of Euler angles

Fig 2.2: Forward kinematic of Euler angles

The rotation matrix will be

R=A*B*C

cos ϕcos θcos𝜓 − 𝑠𝑖𝑛 𝜙𝑠𝑖𝑛𝜓 −cos ϕcos θsin𝜓 − 𝑠𝑖𝑛 𝜙𝑐𝑜𝑠𝜓 cos ϕsin θ
R=[ 𝑠𝑖𝑛𝜓cos θcos𝜓 + cos ϕ 𝑠𝑖𝑛𝜓 −𝑠𝑖𝑛𝜓cos θsin𝜓 + cos ϕ 𝑐𝑜𝑠𝜓 𝑠𝑖𝑛 𝜙sin θ ]
−sin θcos ϕ sin θsin ϕ cos θ
2.2.1 Inverse kinematics of Euler angles
We know the rotational matrix is R
cos ϕcos θcos𝜓 − 𝑠𝑖𝑛 𝜙𝑠𝑖𝑛𝜓 −cos ϕcos θsin𝜓 − 𝑠𝑖𝑛 𝜙𝑐𝑜𝑠𝜓 cos ϕsin θ R11 𝑅12 𝑅13
R=[ 𝑠𝑖𝑛𝜓cos θcos𝜓 + cos ϕ 𝑠𝑖𝑛𝜓 −𝑠𝑖𝑛𝜓cos θsin𝜓 + cos ϕ 𝑐𝑜𝑠𝜓 𝑠𝑖𝑛 𝜙sin θ ]=[𝑅21 𝑅22 𝑅23]
−sin θcos ϕ sin θsin ϕ cos θ 𝑅31 𝑅32 𝑅33
Case 1: If sin θ≠0 then we have
ϕ⁡= atan2(±R23, ±R13)
θ = atan2(√𝑅132 + 𝑅232 ,R33)
𝜓 = atan2(±R32, ±R31)
Case 2: If 𝑅132 + 𝑅232 ≠ 0 then the Euler angles are
ϕ⁡= atan2(±R23, ±R13)
θ = atan2(√𝑅132 + 𝑅232 ,R33)
𝜓 = atan2(±R32, ±R31)
Case 3: If 0< θ < π then we have the following unique Euler angles
ϕ⁡= atan2(±R23, ±R13)
θ = atan2(√𝑅132 + 𝑅232 ,R33)
𝜓 = atan2(±R32, -R31)
Case 4: If 𝑅132 + 𝑅232 = 0 then we have
ϕ = Arbitrary
θ = 90-R33*90
𝜓 = atan2(R21, R22)-R33*⁡ϕ
Case 5: If θ=0 or θ=π then there is infinite number of combinations of ϕ and 𝜓 for just one orientation.
ϕ=0
𝜓 = atan2(R21, R22)
2.3 ROLL, PITCH AND YAW
Euler angles refer to the angles in a sequence of rotations in a body-fixed frame. On the other hand, the roll-pitch-
yaw angles are a sequence of rotations about axes of the space frame. To learn more

Fig: 2.3.1 Roll, pitch and yaw

about the body frame and the space frame, please refer to the lesson on an introduction to configurations.
Visualize Roll-Pitch-Yaw angles as a sequence of rotations relative to the space frame as the following the above
diagram
Visualization of the XYZ roll-pitch-yaw angles. The body frame is initially coincident with the space frame and
then goes through a rotation about the space frame’s x-axis followed by a rotation about the space frame’s y-axis
and finally a rotation about the space frame’s z-axis.
Originated from the following sequence of rotations
1. A rotation of 𝛾 (yaw) about the fixed X axis.
2. A rotation of β (pitch) about the fixed Y axis.
3. A rotation of α (roll) about the fixed Z axis.
Equivalent rotation matrix R= R z (α)* R y (𝛽)* R x (𝛾)
cos 𝛽cos α cos αsin𝛽sin𝛾 − 𝑠𝑖𝑛 α𝑐𝑜𝑠𝛾 cos αsin 𝛽𝑐𝑜𝑠𝛾 + 𝑠𝑖𝑛 α𝑠𝑖𝑛𝛾 R11 𝑅12 𝑅13
R=[ sin αcos 𝛽 𝑠𝑖𝑛αsin 𝛽sin𝛾 + cos α𝑐𝑜𝑠𝛾 𝑠𝑖𝑛 αsin 𝛽𝑐𝑜𝑠𝛾 − cos α 𝑠𝑖𝑛𝛾 ]=[𝑅21 𝑅22 𝑅23]
−sin 𝛽 cos 𝛽sin𝛾 cos 𝛽𝑐𝑜𝑠𝛾 𝑅31 𝑅32 𝑅33
2.3.1 Inverse kinematics of Euler angles
sin⁡𝛽 = -R31
cos 𝛽 = ±√𝑅112 + 𝑅212
Case 1: If cos 𝛽≠0 and 𝛽≠±90 then
𝛽 = atan2(-R31,⁡√𝑅112 + 𝑅212 ) or 𝛽 = atan2(-R31,−√𝑅112 + 𝑅212 )
There fore α and 𝛾 can be calculated by using the following equations
𝑅21 𝑅11
α = atan2( , )
cos 𝛽 cos 𝛽
𝑅32 𝑅33
𝛾 = atan2( , )
cos 𝛽 cos 𝛽

Case 2: If 𝛽 =90 then substitute the value in equation we will get the values as
R31= -1
𝛾-⁡α = atan2(R12, R22)
𝛾-⁡α = atan2(-R23, R13)
When α = 0, 𝛾⁡⁡can be calculated using one of the equations above
Case 3: If 𝛽 =-90 then inserting the value in the equation (6.9.2) we will get the values as
R31= 1
𝛾+α = atan2(-R12, R22)
𝛾+α = atan2(-R23, -R13)
Then the possible solution is α = 0 and 𝛾+α = atan2(-R12, R22)
𝛽=±90 is the singularity of the xyz roll, pitch and yaw angles representation. There for infinitely many sets of roll,
pitch and yaw angles.
3. PROBLEM STATEMENT
3.1 Problem statement and solution
Assume we know the transform 𝐵𝑇𝑇 in the above diagram. which describes the frame at the manipulators finger
tips{T} relative to the base of the manipulator{B}, That we know where the table top is located in space with
respect to manipulator base (because where a description of the frame {S} that is located to the table as
shown,⁡ 𝐵𝑆𝑇 ), and that we know that location of frame attached to the bolt lying on the table with respect to frame
𝑆
𝐺𝑇 .
Figure 3.1: Manipulator reaching for a bolt

Calculate the position and orientation of bolt relative to the manipulators hand 𝐺𝑇 𝑇 .
1. Robot hand without offset distance.
2. Robot hand with offset distance.

We know the 𝐵𝑇𝑇 and we know the table top is located in space with respect to manipulator base{B}, we know
the that location of frame attached to the bolt lying on the table with respect to table frame. So, we know the
transformations of 𝐵𝑇𝑇 , 𝐵𝑆𝑇 and 𝐺𝑆𝑇 .
So, the desired pose is 𝑇
𝐺𝑇 = [( 𝐵𝑇𝑇 )* (𝑇𝑡𝑜𝑜𝑙 )] * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 )

𝑇
𝐺𝑇 = [( 𝐵𝑇𝑇 )]−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
Where ( 𝐺𝑇𝑇 )= Bolt location with respect to tool tip.

( 𝐵𝑇𝑇 )= Tool tip with respect to base frame.

( 𝐵𝑆𝑇 )= Table frame with respect to base frame.
( 𝐺𝑆𝑇 )= Bolt location with respect to table frame.
(𝑇𝑡𝑜𝑜𝑙 )= Tool matrix
cos⁡(𝜃1 − 𝜃2 − 𝜃4 ) sin⁡(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑥
sin⁡(𝜃1 − 𝜃2 − 𝜃4 ) −cos⁡(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑦
( 𝐵𝑇𝑇 )=⁡[ ]
0 0 −1 𝑧
0 0 0 1
Where x=𝑎1 ∗ cos(𝜃1 ) + 𝑎2 ∗ cos⁡(𝜃1 − 𝜃2 )
y=𝑎1 ∗ sin(𝜃1 ) + 𝑎2 + sin⁡(𝜃1 − 𝜃2 )
z=𝑑1 − 𝑑3 − 𝑑4
0 −1 0 𝑡𝑥
1 ⁡0 0 𝑡𝑦
( 𝐵𝑆𝑇 )= [ ]
0 0 ⁡1 𝑡𝑧
0 0 0 1
Where 𝑡𝑥 = Distance along X axis between Base frame and table frame.
𝑡𝑦 = Distance along Y axis between Base frame and table frame.
𝑡𝑧 = Distance along Z axis between Base frame and table frame.
1 0 0 𝑝𝑥
0 ⁡1 0 𝑝𝑦
( 𝐺𝑆𝑇 )= [ ]
0 0 ⁡1 𝑝𝑧
0 0 0 1
Where 𝑝𝑥 = Distance along X axis between Bolt position and table frame.
𝑝𝑦 = Distance along Y axis between Bolt position and table frame.
𝑝𝑧 = Distance along Z axis between Bolt position and table frame.
3.2. Robot hand without offset distance.
𝑇
𝐺𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
The Home position of robot is q= (𝜃1 , 𝜃2 , 𝑑3 , 𝜃3 )=(0,0,100,90)
(𝑎1 ,𝑎2 ) = (425, 375) mm, (𝑑1 , 𝑑4 )=(877, 200) mm (𝑡𝑥 , 𝑡𝑦 , 𝑡𝑧 ) = (10, 3, 7) mm
𝑝𝑥 , 𝑝𝑦 , 𝑝𝑧 = (5, 2, 1) mm
1 0 0 0
0 ⁡1 0 0
(𝑇𝑡𝑜𝑜𝑙 ) = [ ]
0 0 ⁡1 127
0 0 0 1
Now we have to calculate 𝐺𝑇𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
Let’s calculate ( 𝐵𝑇𝑇 )−1 =
−1
cos(𝜃1 − 𝜃2 − 𝜃4 ) sin(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑎1 ∗ cos(𝜃1 ) + 𝑎2 + cos(𝜃1 − 𝜃2 )
) − cos(𝜃1 − 𝜃2 − 𝜃4 ) 𝑎2 ∗ sin(𝜃1 ) + 𝑎2 + sin(𝜃1 − 𝜃2 ) ] =
[ sin(𝜃1 − 𝜃2 − 𝜃4 0
0 0 −1 𝑑1 − 𝑑3 − 𝑑4
0 0 0 1
Substitute all values above mentioned so we will get
 1 0 0 8000 −1 1 0 0 −800 0 −1 0 10
0 ⁡0 −1 0 0 ⁡0 1 −577 1 ⁡0 0 3
( 𝐵𝑇𝑇 )−1 =[ ] , ⁡⁡( 𝐵𝑇𝑇 )−1 = [ ]⁡⁡,⁡⁡⁡⁡⁡⁡⁡( 𝐵𝑆𝑇 ) = [ ]
0 1 0 577 0 −1 0 0 0 0 ⁡1 7
0 0 0 1 0 0 0 1 0 0 0 1
1 0 0 5
𝑆 0 ⁡1 0 2
(𝐺𝑇) = [ ]
0 0 ⁡1 1
0 0 0 1
Substitute above all matrixes in below equation we will get
𝑇
𝐺𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
1 0 0 −800 0 −1 0 10 1 0 0 5 1 0 0 0
𝑇 0 ⁡0 1 −577 1 ⁡0 0 3 0 ⁡1 0 2 0 ⁡1 0 0
𝐺𝑇 =[ ]* [ ]*[ ]*[ ]
0 −1 0 0 0 0 ⁡1 7 0 0 ⁡1 1 0 0 ⁡1 127
0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1
0 −1 0 −792
𝑇 0 ⁡0 1 −442
𝐺𝑇 = [ ]
−1 0 0 −8
0 0 0 1
Final position of Tool tip = P (X, Y, Z) = (-792, -442, -8)
3.2.1 Euler angles
0 −1 0 R11 𝑅12 𝑅13
The rotational matric R=[ 0 0 1]=[𝑅21 𝑅22 𝑅23]
−1 0 0 𝑅31 𝑅32 𝑅33
The Euler angles can be calculated as fallows
ϕ⁡= atan2(±R23, ±R13)
θ = atan2(√𝑅132 + 𝑅232 ,R33)
𝜓 = atan2(±R32, ±R31)
If 𝑅132 + 𝑅232 ≠ 0 then the Euler angles are
ϕ⁡= atan2(±1, ±0) = 90 or -90 degrees
θ = atan2(√02 + 12 ,0) = 90 or -90 degrees

𝜓 = atan2(±0, ± (-1)) = 0 degree

Therefore, the Euler angles are (ϕ, θ, 𝜓) = (90, 90, 0) or (-90, -90, 0) degrees
3.2.2 Roll, Pitch and yaw ( 𝛂, 𝜷,⁡ 𝜸)
cos 𝛽cos α cos αsin𝛽sin𝛾 − 𝑠𝑖𝑛 α𝑐𝑜𝑠𝛾 cos αsin 𝛽𝑐𝑜𝑠𝛾 + 𝑠𝑖𝑛 α𝑠𝑖𝑛𝛾
The rotational matric R= [ sin αcos 𝛽 𝑠𝑖𝑛αsin 𝛽sin𝛾 + cos α𝑐𝑜𝑠𝛾 𝑠𝑖𝑛 αsin 𝛽𝑐𝑜𝑠𝛾 − cos α 𝑠𝑖𝑛𝛾 ] =
−sin 𝛽 cos 𝛽sin𝛾 cos 𝛽𝑐𝑜𝑠𝛾
0 −1 0 R11 𝑅12 𝑅13
[0 0 1] = [𝑅21 𝑅22 𝑅23]
−1 0 0 𝑅31 𝑅32 𝑅33
If cos 𝛽 = ±√𝑅112 + 𝑅212 = cos 𝛽= 0 && 𝛽 =90
R31= -1, sin 𝛽 = -1, 𝛽 = -90
if 𝛽⁡= 90 then rotational matric will be
0 sin⁡(𝛾 − ⁡α ) cos⁡(𝛾 − ⁡α ) 0 −1 0
[ 0 cos⁡(𝛾 − ⁡α ) −sin⁡(𝛾 − ⁡α )] = [ 0 0 1]
−1 0 0 −1 0 0
𝛾-⁡α = atan2(R12, R22)
𝛾-⁡α = atan2(-R23, R13)
When α = 0, 𝛾⁡⁡can be calculated using one of the equations above
𝛾 = atan2(-1,0) = -90
 if 𝛽⁡= -90 then rotational matric will be
0 −sin⁡(𝛾 + ⁡α ) −cos⁡(𝛾 + ⁡α ) 0 −1 0
[0 cos⁡(𝛾 + ⁡α ) −sin⁡(𝛾 + ⁡α ) ] = [ 0 0 1]
−1 0 0 −1 0 0
𝛾-⁡α = atan2(R12, R22)
𝛾-⁡α = atan2(-R23, R13)
When α = 0, 𝛾⁡⁡can be calculated using one of the equations above
𝛾 = atan2(1,0) = 90
Therefore, roll pitch and yaw ( α, 𝛽,⁡ 𝛾) = (0, 90, -90) or (0, -90, 90)
3.3. Robot hand with offset distance.
𝑇
𝐺𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
The Home position of robot is q= (𝜃1 , 𝜃2 , 𝑑3 , 𝜃3 )=(0,0,100,90)
(𝑎1 ,𝑎2 ) = (425, 375) mm, (𝑑1 , 𝑑4 )=(877, 200) mm (𝑡𝑥 , 𝑡𝑦 , 𝑡𝑧 ) = (10, 3, 7) mm
𝑝𝑥 , 𝑝𝑦 , 𝑝𝑧 = (5, 2, 1) mm
cos⁡(60) 0 sin⁡(60) 7.5"
0 ⁡1 0 0
(𝑇𝑡𝑜𝑜𝑙 )= [ ]
−sin⁡(60) 0 ⁡cos⁡(60) 6"
0 0 0 1
1inch=25.4mm

Figure 6.2: Tool with offset distance

0.5 0 0.86 190.5

0 ⁡1 0 0
(𝑇𝑡𝑜𝑜𝑙 )= [ ]
−0.86 0 ⁡0.5 152.4
0 0 0 1
Now we have to calculate 𝐺𝑇𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) *(𝑇𝑡𝑜𝑜𝑙 )
Let’s calculate ( 𝐵𝑇𝑇 )−1 =
−1
cos(𝜃1 − 𝜃2 − 𝜃4 ) sin(𝜃1 − 𝜃2 − 𝜃4 ) 0 𝑎1 ∗ cos(𝜃1 ) + 𝑎2 + cos(𝜃1 − 𝜃2 )
) − cos(𝜃1 − 𝜃2 − 𝜃4 ) 𝑎2 ∗ sin(𝜃1 ) + 𝑎2 + sin(𝜃1 − 𝜃2 ) ]
[ sin(𝜃1 − 𝜃2 − 𝜃4 0
0 0 −1 𝑑1 − 𝑑3
0 0 0 1

Substitute all values above mentioned so we will get

 1 0 0 800 −1 1 0 0 −800 0 −1 0 10
0 ⁡0 −1 0 0 ⁡0 1 −577 1 ⁡0 0 3
( 𝐵𝑇𝑇 )−1 = [ ] , ( 𝐵𝑇𝑇 )−1 = [ ]⁡⁡⁡⁡⁡⁡⁡,⁡( 𝐵𝑆𝑇 )= [ ]
0 1 0 577 0 −1 0 0 0 0 ⁡1 7
0 0 0 1 0 0 0 1 0 0 0 1
1 0 0 5
𝑆 0 ⁡1 0 2
( 𝐺 𝑇 )= [ ]
0 0 ⁡1 1
0 0 0 1
Substitute above all matrixes in below equation we will get
𝑇
𝐺𝑇 = ( 𝐵𝑇𝑇 )−1 * ( 𝐵𝑆𝑇 )* ( 𝐺𝑆𝑇 ) ∗ (𝑇𝑡𝑜𝑜𝑙 )
1 0 0 −800 0 −1 0 10 1 0 0 5 0.5 0 0.86 190.5
𝑇 0 ⁡0 1 −577 1 ⁡0 0 3 0 ⁡1 0 2 0 ⁡1 0 0
𝐺𝑇 = [ ]*[ ]*[ ]*[ ]
0 −1 0 0 0 0 ⁡1 7 0 0 ⁡1 1 −0.86 0 ⁡0.5 152.4
0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1
0 1 0 −792
𝑇 −0.86 0 0.5 −416.6
𝐺𝑇 =[ ]
−0.5 0 −0.86 −198.5
0 0 0 1
Final position of Tool tip = P (X, Y, Z) = (-792, -416.6, -198.5)
3.3.1 Euler angles
0 −1 0 R11 𝑅12 𝑅13
The rotational matric R=[−0.86 0 0.5 ]=[𝑅21 𝑅22 𝑅23]
−0.5 0 −0.86 𝑅31 𝑅32 𝑅33
The Euler angles can be calculated as fallows
ϕ⁡= atan2(±R23, ±R13)
θ = atan2(√𝑅132 + 𝑅232 ,R33)
𝜓 = atan2(±R32, ±R31)
If 𝑅132 + 𝑅232 ≠ 0 then the Euler angles are
ϕ⁡= atan2(±0.5, ±0) = 90 or -90 degrees
θ = atan2(√02 + 0.52 ,-0.86) = -30.17 or 30.17 degrees
𝜓 = atan2(±0, ± (-0.5)) = 0 degree
Therefore, the Euler angles are (ϕ, θ, 𝜓) = (90, -30.17, 0) or (-90, 30.17, 0) degrees
3.3.2 Roll, Pitch and yaw ( 𝛂, 𝜷,⁡ 𝜸)
cos 𝛽cos α cos αsin𝛽sin𝛾 − 𝑠𝑖𝑛 α𝑐𝑜𝑠𝛾 cos αsin 𝛽𝑐𝑜𝑠𝛾 + 𝑠𝑖𝑛 α𝑠𝑖𝑛𝛾
The rotational matric R= [ sin αcos 𝛽 𝑠𝑖𝑛αsin 𝛽sin𝛾 + cos α𝑐𝑜𝑠𝛾 𝑠𝑖𝑛 αsin 𝛽𝑐𝑜𝑠𝛾 − cos α 𝑠𝑖𝑛𝛾 ] =
−sin 𝛽 cos 𝛽sin𝛾 cos 𝛽𝑐𝑜𝑠𝛾
0 −1 0 R11 𝑅12 𝑅13
[−0.86 0 0.5 ] = [𝑅21 𝑅22 𝑅23]
−0.5 0 −0.86 𝑅31 𝑅32 𝑅33
√3
If cos 𝛽 = ±√𝑅112 + 𝑅212 = cos 𝛽= && 𝛽 ≠ ±⁡90 then 𝛽⁡value can be
2

𝛽 = atan2(-R31,±√𝑅112 + 𝑅212 )
𝛽 = atan2(0.5,±√02 + −0.862 ) = 30.17 or -30.17 Degrees
Cos 𝛽 = 0.86
There fore α and 𝛾 can be calculated by using the following equations
𝑅21 𝑅11 −0.86 0 −0.86 0
α = atan2( , ) = atan2( , ) or atan2( , ) = -90 or 90 Degrees
cos 𝛽 cos 𝛽 0.86 0.86 −0.86 −0.86
𝑅32 𝑅33 0 −0.86 0 −0.86
𝛾 = atan2( , ) = atan2( , ) or atan2( , )=0
cos 𝛽 cos 𝛽 0.86 0.86 −0.86 −0.86

Therefore, roll pitch and yaw ( α, 𝛽,⁡ 𝛾) = (-90, 30.17, 0) or (90, -30.17, 0)
 4. RESULTS

For the selected problem of this project a structural design and assembly of suitable gripper has been designed for
plucking pineapple from pine apple plant as shown figure below.

Apporach of gripper in such a way that it can suitable to hold the pine apple and twist opetion to pluck the
pineapple.As there is two,three and multi fingure gripper available but two finger gripper feasible for hold and

twist opertation. Studied about the clamping force, torque, payload of grippe and power required for the motor.In
this design of gripper individual compoents has been designed and assembly has been done by using solid works
software. Suitable material for the gripper has been done selected. Theoritical study of kinematic analysis for
gripper has been done.

The forward kinematics analsys of scara robot arm of with out offset distance and with offset distance of tool, The
final postion of tool tip has been calcualted accoding to refernce of the robot Tool{T} to the bolt postion{G}.
clearly we can achive bolt postion with respect to tool postion, with the help of tool postion withrespect to base and
bolt postion with respect to table postion. The joint variables of the scara robot as given input values so, that the
final position of tool tipe has been obtained as fallowing below.

4.1 Forward kinematics of scara robot without offset distance having tool :

Robot hand without offset distance of tool, the joint variables are the input values so, That we can achive the final
postion of homogenous transform matix bolt with respect to the tool tip. The final postion of tool tip values are
shown below in table 7.1.

Table 4.1: Robot hand with out offset distance

Joint Variables Position vector Euler angles Roll Yaw Pitch

(ϕ, θ, 𝜓)degrees ( α, 𝛽,⁡ 𝛾) degrees
𝜃1 = 0
𝜃2 = 0 P(X)= (−792) (90, 90, 0) (0, 90, -90)
𝑑3 = 100 P(Y)= (-442) Or Or
𝜃3 = 90 P(Z)= (-8) (-90, -90, 0) (0, -90, 90)

4.2 Forward kinematics of scara robot with offset distance having tool :

Robot hand with offset distance of tool, the joint variables are the input values so, That we can achive the final
postion of homogenous transform matix bolt with respect to the tool tip. The final postion of tool tip values are
shown below in table 7.2.
 Table 4.2: Robot hand offset distance

Joint Variables Position vector Euler angles Roll Yaw Pitch

(ϕ, θ, 𝜓)degrees ( α, 𝛽,⁡ 𝛾) degrees
𝜃1 = 0
𝜃2 = 0 P(X)= (−792) (90, -30.16, 0) (-90, 30.17, 0)
𝑑3 = 100 P(Y)= (-416.6) Or Or
𝜃3 = 90 P(Z)= (-198.5) (-90, 30.17, 0) (90, -30.17, 0)

5. CONCLUSIONS
In this work an attempt has been made to design and to perform kinematic analysis of a two – fingered parallel type
gripper or an end – effector for pineapple plucking task.
Due to the presence of sharp edges of leaf’s in the pineapple plant, employing robot hand in place of human hand
for plucking of pineapples from pineapple plant would result in speedy removal of fruits with ease and without any
injury to the human worker. And without damaging the plant as such.
Design of pineapple gripper has been done in such a way that it is suitable for holding and twisting operation of
pineapple to pluck pineapple from pineapple plant. It is also proposed to have cylindrical shape for the interior of
fingers or jaws so that gripper can handle the pineapple with predefined force without any slippage. Moreover, load
carrying capacity of the gripper is such that it can grasp or hold any size of the pineapple without any damage to
peripheral surface of the pineapple.
This implies that the gripper design as proposed in the present study is feasible from the point of view of selected
or stated design criteria.
A detailed design and the subsequent assembly of the components of the parallel jaw gripper have been carried out
by utilizing solid works software, version 2020.
In this work possible and feasible solutions of forward kinematic analyses of Scara robot has been done as
fallowing

REFERENCES

[1] Robert J Shilling “Fundamentals of robotics analysis and control” (2001).

[2] Ya Xiong, Yuanyue Ge, Pal Johan from (2020) Faculty of Science and Technology, Norwegian University of
Life Sciences.

[3] A. Gongal, S.Amatya, M. Karkee a, Q. Zhang b, K. Lewis (2015) Biological Systems Engineering Department,
Washington State University, United States, Center for Precision and Automated Agricultural Systems,
Washington State University, United

[4] Yuanshen Zhao, Liang Gong, Yixiang Huang, Chengliang Liu (2016) School of Mechanical Engineering,

[5] Mingyou Chen, Yunchao Tang, Xiangjun Zou, Zhaofeng Huang, Hao Zhou, Siyu Chen(2021) Key Laboratory
of Key Technology on Agricultural Machine and Equipment, College of Engineering, South China Agricultural
University, Guangzhou 510642, China.
[6] Lufeng Luoa, Yunchao Tangb, Qinghua Lua, Xiong Chenc, Po Zhangc, Xiangjun Zouc (2018) College of
Mechanical and Electrical Engineering, Foshan University

[7] Shivaji Bachche (2015) Institute of Multidisciplinary Research for Advanced Materials, Tohoku University,
Katahira.

[8] Nischit GS Published on topic of Design of a Smart Gripper for Collaborative Robots in International Journal
of Engineering Research & Technology (IJERT) ISSN: 2278-0181.

[9 A F Guo1, J Li1, L Q Guo1, T Jiang1 and Y P Zhao (2020) School of Mechanical and Automotive Engineering,
Liaocheng University, Liaocheng, 252000 China.