UNIVERSITI TEKNOLOGI MALAYSIA

FAKULTI KEJURUTERAAN ELEKTRIK

SKEM 4143 : ROBOTICS

Semester I, 2015/2016 Academic Session

ASSIGNMENT

No. Assignment No. Problems Due Date

1. Assignment Set 1 a. SKEM /SEM 4143 2014/2015-I Test 1 :

DO ALL QUESTIONS

b. SKEM /SEM 4143 2014/2015-I Final :

DO Q1 & Q2

2. Assignment Set 2 a. SKEM /SEM 4143 2014/2015-I Test 2 :

DO ALL QUESTIONS

b. SKEM /SEM 4143 2014/2015-I Final :

DO Q3 & Q4
 ROBEEX120142015I.DOC DrJ2014

Name :
UNIVERSITI TEKNOLOGI MALAYSIA
FAKULTI KEJURUTERAAN ELEKTRIK
SEMESTER : 2014/2015 - I
Student ID No. : SEM 4143 : ROBOTICS
TEST 1
PART A (50%) :
1. Figure 1 illustrates a FIVE (5) degree-of-freedom Fanuc M100 robot developed by Fanuc Robotics America Inc.
for inductrial applications. By using the Denavit-Hartenberg representation, establish the orthonormal link
coordinate systems (please clearly sketch on the diagram supplied) and derive the corresponding D-H parameters
for all the joints as accurately as posible from the given data. [50 marks]

+ +
+

+
+

Joint i i di ai

Figure 1 : A Five dof Fanuc M100

Robot
ROBEEX120142015I.DOC DrJ2014

PART B (15%) :

1. List down at least five (5) reasons the advantages of using industrial robots. [5 marks]

Ans : .....

......

......

......

...

2. Briefly describes the following types of transmission mechanisms in robotic arm : [5 marks]

Spur gears : Ans : ......

......

Worm gears : Ans : . .....

......

Torsion shafts : Ans : . .... .

......

Toothed rack-and-pinion : Ans : ..

......

3. Briefly describes the following types of robots. [5 marks]

Parallel Robot : Ans : .....

.....

Direct Drive Robot : Ans : ......

.....

Flexible-Link Robot : Ans : ......

......

Point-to-point Control Robot : Ans : ....

......
ROBEEX120142015I.DOC DrJ2014

PART C (35%) :

1. Consider the 3D shape pictured in Figure 2 below. Find the transformation matrix T10 . The
front and back faces are regular hexagons, the length of each side is 2 3 . The six side faces
are all identical rectangles that have two sides with length 2 3 and two sides with length 5.
The interior angles of a regular hexagon are 120 . [20 marks]

Figure 2 : Coordinate Frames for

Question 1 Part C

Ans : ..
ROBEEX120142015I.DOC DrJ2014

2. The ox0 y0 z0 frame is a base coordinates while the ox1 y1z1 frame is a rotating coordinate frame.
Lets P0 2 4 6 be a point P relative to the reference frame ox0 y0 z0 . Determine the
T

resultant location of the point P relative to the reference frame ox0 y0 z0 after undergoing the
following series of rotations :

1. A rotation with an angle 90 about the oz0 axis ( Rz0 ,90 )

2. Followed by a rotation with an angle 90 about the oy1 axis ( R y1 ,90 )
3. Followed by a rotation with an angle 90 about the oz1axis ( Rz1 ,90 ).

Ans : [15 marks]

ROBEEX220142015I.DOC

UNIVERSITI TEKNOLOGI MALAYSIA

FAKULTI KEJURUTERAAN ELEKTRIK
SEMESTER : 2014/2015 - I
SEM 4143 : ROBOTICS
TEST 2

1. Figure 1 illustrates a 2 dof Cartesian robot. m1 and m2 , m3 are the mass for link 1
and 2, respectively. l1 and l2 are the length for link 1 and 2, respectively. d1 and d 2
are the joint variable for joint 1 and 2, respectively.

The homogeneous transformation matrices and the inertia matrices are :

1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 ; 0
1 0 0; 0 0 0 ; 0 0 0 0 .
T01 T1
2
J1 J2
0 0 1 d1 0 0 1 d2 0 0 0 0 0 0 0 0

0 0 0 1 0 0 0 1 0 0 0 m1 0 0 0 m2

The inertia tensor of link ith about its center of mass with respect to frame ith is :

m l2 m l2
D1 1 1 diag 1, 1, 0 ; D2 2 2 diag 1, 1, 0 .
12 12

The center of mass for each link with respect to its coordinate frame ith is as follows (Note :
ci ri ) :

T T
l l
c1 0 , 0 , 1 , 1 ; c2 0 , 0 , 2 , 1 ;
2 2

Derive the robots dynamic equations. [50 marks]

d2

m2

m1 d1
FIGURE 1 : A 2 dof
Cartesian Robot

10/14/2015 1 DrJ2012
ROBEEX220142015I.DOC

2. Consider the joint interpolated trajectory design problem as shown in Figure 2. It is

desired that the robots joint (t ) is to start from position (t0 ) (0) 1 with zero
velocity and move as fast as possible to (t f ) (9) 3 without hitting the obstacle.
This may be achieved by specifying the knot points to go around the obstacle as
shown. The robot does not stop at the pre-assigned knot points. The robot stops
only when it reaches the final position (9) 3 .

Fully determine the joint interpolated trajectory if a second order polynomials

(t ) At 2 Bt C are being used for the trajectory from the starting position to the
knot point 1, and between knot point 1 to knot point 2, and from the knot point 2
to the final position. [50
marks]

FIGURE 2

10/14/2015 2 DrJ2012
 CONFIDENTIAL

UNIVERSITI TEKNOLOGI MALAYSIA

FINAL EXAMINATION
SEMESTER 2014/2015 - I

SUBJECT CODE : SKEM 4143

SUBJECT : ROBOTICS

LECTURER : PROF. DR. JOHARI HALIM SHAH OSMAN

COURSE : SEM

TIME : 2 HOURS 30 MINUTES

DATE :

INSTRUCTION TO CANDIDATES :

i ) ANSWER ANY FOUR (4) QUESTIONS.

ii) PLEASE ATTACH PAGE 12 TOGETHER WITH THE ANSWER SCRIPT

BOOKLET IF YOU ANSWER QUESTION NO 2.

THIS EXAMINATION BOOKLET CONSISTS OF 12 PAGES INCLUDING THE FRONT COVER

 -2
SKEM/SEM 4143
QUESTION 1 :

Figure 1 shows a rigid rectangular block where a coordinate frame o1 x1 y1 z1 is attached to.
The block is initially positioned in space with reference to the coordinate frames o0 x0 y0 z0
and o2 x2 y2 z2 which are fixed frames of reference.

A) Express the homogeneous transformation matrix of the o1 x1 y1 z1 coordinate frame

with respect to the o0 x0 y0 z0 frame, T01 . [3 marks]

B) Express the homogeneous transformation matrix of the o2 x2 y2 z2 coordinate frame

with respect to the o0 x0 y0 z0 frame, T02 . [3 marks]

C) Express the homogeneous transformation matrix of the o1 x1 y1 z1 coordinate frame

with respect to the o2 x2 y2 z2 frame, T21 . [3 marks]

D) The block then undergoes the following ordered sequence of motion :

a. a rotation of 30 anti-clockwise about the z0 -axis, followed by

b. a rotation of 45 anti-clockwise about the y0 -axis, followed by
c. a rotation of 90 anti-clockwise about the x1 -axis, followed by
d. a rotation of 60 anti-clockwise about the x2 -axis.

i) Determine the homogeneous transformation matrix for the new position and
orientation of coordinate frame o1 x1 y1 z1 with respect to the o0 x0 y0 z0 frame after
the four motions. [13 marks]

ii) Determine the homogeneous transformation matrix for the new coordinates of
the vertex point C with respect to the o0 x0 y0 z0 frame after the four motions.
[3 marks]

z0 x2
o1

3 y1
x1
C
10 2 6
4
o0 y0 z2

FIGURE 1 : A Set Of Three Coordinates Frames For Question 1

DrJ-2014.
 -3
SKEM/SEM 4143
QUESTION 2 :

Motoman SIA20D developed by Yaskawa is an agile and versatile 7-axis compact, lean
and powerful arm that can be used for a wide range of industrial applications such as
product assembly, injection molding, inspection, and machine tending. Figure 2 below
shows a picture of the robot, its schematic diagram and dimension (mm).

By using Denavit-Hartenberg (D-H) representation and standard convention, as accurately

as possible:

A) Establish the orthonormal link coordinate systems for all the joints (please clearly
sketch on the schematic diagram supplied on Page 12); [12 marks]

B) Derive the corresponding D-H parameters for all the joints. [11 marks]

C) Express the robots forward kinematics in terms of the homogeneous

transformation matrices Ti i 1 . (You do not need to simplify your answer. Express
your answer in terms of matrix products expression.) [2 marks]

Side View Back View

FIGURE 2 : Motoman SIA20D Robot For Question 2

DrJ-2014.
 -4
SKEM/SEM 4143
QUESTION 3 :

A) Newton-Euler formulation of the dynamic model of a manipulator is computationally

more efficient than the Lagrange-Euler formulation. However, most researchers still
use the Lagrange-Euler formulation. Why and explain (Give at least two reasons).
[5 marks]

B) Figure 3 illustrates a Microbo ECUREUIL robot developed by a Swiss firm. The

homogeneous transformation matrices, moments of inertia and pseudo inertia matrices
are, if needed, are as given below :

Cos1 - Sin1 0 0 0 0 1 0 1 0 0 0
Sin Cos 0 0 , 1 0 0 0 , 0 1 0 0 , 1
T0
1 1 1
T1
2
T2
3 Ii mi li2
0 0 1 0 0 1 0 d2 0 0 1 d3 3

0 0 0 1 0 0 0 1 0 0 0 1

13 m1l12 0 0 1 2 m1l1 0 0 0 0 0 0 0 0
0
0 0 0 0 , 0 0 0 0 , 0 0 0 , i 1, 2, 3
J1 J2 J3
0 0 0 0 0 0 0 0 0 0 0 0
1
2 m1l1 0 0 m1 0 0 0 m2 0 0 0 m3

Using either the Lagrange-Euler or the Newton-Euler method, determine the effect
of the Coriolis and centrifugal forces on the robot dynamics. [20 marks]

l3
m2 d2 l1
d3
m3

5 m1
4

FIGURE 3 : Microbo ECUREUIL Robot For Question 3

DrJ-2014.
 -5
SKEM/SEM 4143

QUESTION 4 :

A) Explain the difference between the Joint Trajectory Planning to the Cartesian
Trajectory Planning, their advantages and disadvantages, and why one is preferred as
compared to the other for a certain type of application. [6 marks]

B) A robots joint is required to start from rest at position 30 at t0 0 sec and move
to the final position at 60 . The robot stops when it reaches the position
60 . Let the joint motor has an acceleration constraint of a 20
(degree/sec2) and a velocity constraint of v 5 (degree/sec).

i) Design the robot joint trajectory (t ) in terms of t using the Linear Segments
with Parabolic Blends (LSPB) trajectory planning method.
[10 marks]

ii) Evaluate the possibility of using the Bang-Bang Parabolic Blends (BBPB)
trajectory planning method to design the desired joint trajectory. [9 marks]

DrJ-2014.
 -6
SKEM/SEM 4143

QUESTION 5 :

A) With the aid of sketches, briefly explain the basic operation AND an example of
application in robotic arm/system for ANY TWO of the following sensors: [8 marks]

i. Ultrasonic Proximity Sensors

ii. Inductive Proximity Sensors
iii. Piezoelectric Force Sensors

B) Figure 4 illustrates the essential features of a system for controlling paper thickness
with the aid of a sensor. The paper is in a wet fiber suspension and is passed between
rollers. By varying the roller separation, paper thickness is regulated. The thickness of
the paper is measured using a sensor placed at the output end of the roller as shown.
The sensor is installed in such a way that a sensing mechanism touches the paper.
Once the measurement has been made, the computer compares with the desired
thickness setpoint. If the thickness of the paper does not satisfy a given tolerance, a
corrective signal will be sent by the computer to the worm screw motor controlling the
movable roller such that the desired thickness is attained.

i) As the engineer for the plant, you are required to identify the appropriate sensor
to be used in measuring the thickness of the paper. Why do you choose that
particular sensor for this application ? [7 marks]

ii) With the aid of sketches, briefly describes the operation of the sensor that you
have chosen. [10 marks]

SENSOR

FIGURE 4 : A Paper-Thickness-Controlling System For Question 5

DrJ-2014.
 -7
SKEM/SEM 4143
APPENDIX

The transformation matrix of ox1 y1z1 coordinates frame with respect to the ox0 y0 z0 coordinates frame :

i1 i0 j1 i0 k1 i0
R01 i1 j0 j1 j0 k1 j0
i1 k0 j1 k0 k1 k0

The basic homogeneous rotation matrices :

1 0 0 0 cos 0 sin 0 cos sin 0 0

0 cos sin 0 0 1 0 0 sin cos 0 0
Tx, , Ty, , Tz,
0 sin cos 0 sin 0 cos 0 0 0 1 0
0 0 0 1 0 0 0 1 0 0 0 1

The basic homogeneous translation matrices :

1 0 0 dx 1 0 0 0 1 0 0 0
0 1 0 0 0 1 0 dy 0 1 0 0
Ttran x ,dx , Ttran y ,dy , Ttranz ,dz
0 0 1 0 0 0 1 0 0 0 1 dz
0 0 0 1 0 0 0 1 0 0 0 1

The homogeneous transformation matrix and its inverse :

n x sx ax dx
n R33 | p 31
s y a y d y n s a p
T01 y |
nz s z a z d z 0 0 0 1
0 013 | 111
0 0 1
n x n y n z n T p
R | R T p
T
s
T sz s p
T
1 1 s
T10 x y
|
a x a y a z a p
0 T

0 0 0 | 111
0 1 13

THE D-H REPRESENTATION (Figure 5)

Step 1 : Establish the base frame. The z0 axis is located on the rotation or translation axis for the joint 1. The
x0 and y0 axes are chosen conveniently to form a right-hand frame.

Step 2 : For each i , i 1, 2, , N 1 , perform Step 3 to 6.

Step 3 : Establish zi axis. Align the zi axis with the axis of rotation or translation (sliding) for joint i 1 .

Step 4 : Locate the origin, oi , for the i th coordinate frame at the intersection of the common normal between
the zi - zi 1 axes and the zi axis. If the zi axis intersects the zi 1 axis, then place oi at the
intersection.

DrJ-2014.
 -8
SKEM/SEM 4143
Step 5 : Establish xi along the common normal between zi 1 and zi through oi , or in the direction normal to
the zi 1 zi plane if zi 1 and zi intersect.

Step 6 : Establish yi to complete a right-hand frame.

Step 7 : Establish the end-effecter frame oN x N y N zN . Assuming the N th joint is revolute, set z N in the
direction of approach for the end-effecter on an object. Establish the origin o N conveniently along
z N , preferably at the center of the gripper or at the tip of any tool carried by the manipulator. Set y N
in the direction of gripper closure, and set x N to complete the right-hand coordinate system. If the
tool is not a gripper, then set x N and y N to form a right-hand coordinate system.

Step 8 : Complete the table for link parameter (D-H parameter) - ai , di , i , i :

ai = common normal distance along xi starting from the intersection of the xi and zi 1 axes to the
origin of the ith coordinate frame, oi . Note that ai is also known as the length of the ith
link.
di = distance along the zi 1 axis starting from the origin for the (i 1) th coordinate frame, oi1 ,
to the intersection of the xi and zi 1 axes. (The distance between two common normal ai
and ai1 , or the distance between the xi and xi1 axes.) Note that d i is variable if joint ith
is prismatic.
i = the twist angle starting from the zi 1 axis to the zi axis, measured about the xi axis. For most
manipulator, the twist angle is in multiples of 90 .
i = rotation angle starting from the xi1 axis to the xi axis, measured about the zi 1 axis. Note
that i is variable if joint i is revolute.

i
Step 9 : Determine the D-H homogeneous transformation matrices Ti 1 by substituting the D-H parameters
into the following equation, where S i and C i are Sin( i ) and Cosine( i ) , respectively :

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
Ti i 1 i .
0 S i C i di
0
0 0 1

FIGURE 5 : Denavit-Hartenberg Coordinate Assignment

DrJ-2014.
 -9
SKEM/SEM 4143
LAGRANGE-EULER EQUATION

The following outlines the Lagrange-Euler equations for deriving the dynamic model of an N dof robot.

N N N
i Dik qk + Cikm qk qm Gi , i 1, 2 ,, N ,
k 1 k 1 m 1

where

N
Dik
j max( i , k )
Tr U jk J j U Tji , i , k 1, 2 ,, N ,

N
Cikm Tr U jkm J j U Tji
j max( i , k , m)
, i , k , m 1, 2 ,, N ,

N
Gi = mj g U ji rj , i 1, 2 ,, N .
j i

T0i T0j 1 Q j T ji1 for j i

U ij , i 1, 2 ,, N .
q j 0 for j i

T0 j 1 Q j T jk11 Qk Tki1 if ik j
U ij k 1
U ijk T0 Qk Tk j11 Q j T ji1 if i j k
qk
0 if i j or i k

0 1 0 0 0 0 0 0
1 0 0 0 0 0 0 0
Qi if the i th joint is revolute, or Qi if the i th joint is prismatic
0 0 0 0 0 0 0 1

0 0 0 0 0 0 0 0

g g x , g y , gz , 0 .
OR

The Lagrange-Euler Equation of Motion is given as follows :

d L L
i , i = 1, 2, ..., N.
dt qi qi

L KP
where K is the total kinetic energy of the robot, P is the total potential energy of the robot, i is the
force or torque applied to the system, and qi is the generalized coordinate of the robotic arm. For
revolute joint, qi i . For prismatic joint, qi di .

DrJ-2014.
 - 10
SKEM/SEM 4143
NEWTON-EULER EQUATION

The following outlines the algorithm for the recursive Newton-Euler formulation of a dynamic model of
an N dof robot.

Step 1 : Set the value T00 I , f N 1 f Tool , nN 1 nTool , 0 0,

0 0, v0 0 ,
v0 g , k 1, and

1 if the ith joint is revolute

i
0 if the ith joint is prismatic

Step 2 : Compute

zk 1 R0k 1 i3

k k 1 k qk zk 1

k k 1 k qk zk 1 k 1 qk zk 1

T0k T0k 1 Tkk1

1
H I 0 , = non-zero scalar factor, normally = 1 in robotic.
3 4

sk H1 T0k T0k 1 i4

vk vk 1 k sk k k sk 1 k qk zk 1 2 k q k zk 1

Step 3 : Set k k 1. If k N , go to Step 2; otherwise set k N and continue.

Step 4 : Calculate :

rk H1 T0k ck i4

f k f k 1 mk vk k rk k k rk

T
Dk R0k Dk R0k

nk nk 1 sk rk f k rk f k 1 Dk k k Dk k

k k nk zk 1 1 k f k zk 1 bk q k
T T

Step 5 : Set k k 1. If k 1, go to Step 4.

DrJ-2014.
 - 11
SKEM/SEM 4143
LINEAR SEGMENT WITH PARABOLIC BLENDS (LSPB) and
BANG-BANG PARABOLIC BLENDS (BBPB) TRAJECTORY

(t)

2
f + 0
3
2
1
0

t t (second)
0 t t t -t t
b f f b f
2

(t0 ) 0 , (t f ) f , (t0 ) (t f ) 0

TIME SEGMENT LSPB TRAJECTORY BBPB TRAJECTORY

First part : 0 t tb 1 (t ) 0
V 2
t 1 (t ) 0
V 2
t
2tb 2tb

V V
1 (t ) t at 1 (t ) t at
tb tb

V V
1 (t ) a 1 (t ) a
tb tb

Second part : tb t t f tb f 0 Vt f
2 (t ) Vt - not applicable, since
2
2 (t ) V t tb t f tb .
2 (t ) 0

Third part : t f tb t t f . 3 (t ) f
at 2f
at f t
a 2
t 3 (t ) f
at 2f
at f t
a 2
t
2 2 2 2
V V
3 (t ) (t f t ) a (t f t ) 3 (t ) (t f t ) a (t f t )
tb tb

V V
3 (t ) a 3 (t ) a
tb tb

Time V 0 f Vt f tf f 0
tb tb
a V 2 a
f 0 V
tf f 0
V a t f 2tb 2
a

DrJ-2014.
 - 12
SKEM/SEM 4143

NAME :

FIGURE 6 : Motoman SIA20D Robotic Manipulator For Question 2

DrJ-2014.